vecId
stringlengths
12
23
id
stringlengths
2
13
conference
stringclasses
11 values
year
float64
2.02k
2.03k
title
stringlengths
6
189
abstract
stringlengths
10
4.74k
author
stringlengths
0
7.45k
aff
stringlengths
0
7.16k
status
stringclasses
11 values
track
stringclasses
4 values
keywords
stringlengths
0
804
github
stringlengths
0
141
site
stringlengths
0
193
gsCitation
float64
-1
11.1k
arxiv
stringlengths
0
12
text
stringlengths
58
4.82k
vector
list
wacv_2025_c95eb6df7b
c95eb6df7b
wacv
2,025
Towards Unbiased Continual Learning: Avoiding Forgetting in the Presence of Spurious Correlations
Continual Learning (CL) has emerged as a paramount area in Artificial Intelligence (AI) because of its ability to learn multiple tasks sequentially without significant performance degradation. Despite the growing interest in CL frameworks a critical aspect must be addressed: the inherent biases within training data. In...
Giacomo Capitani; Lorenzo Bonicelli; Angelo Porrello; Federico Bolelli; Simone Calderara; Elisa Ficarra
Universit `a degli Studi di Modena e Reggio Emilia, Italy; Universit `a degli Studi di Modena e Reggio Emilia, Italy; Universit `a degli Studi di Modena e Reggio Emilia, Italy; Universit `a degli Studi di Modena e Reggio Emilia, Italy; Universit `a degli Studi di Modena e Reggio Emilia, Italy; Universit `a degli Studi ...
Poster
main
https://github.com/aimagelab/mammoth
https://openaccess.thecvf.com/content/WACV2025/html/Capitani_Towards_Unbiased_Continual_Learning_Avoiding_Forgetting_in_the_Presence_of_WACV_2025_paper.html
0
Towards Unbiased Continual Learning: Avoiding Forgetting in the Presence of Spurious Correlations Continual Learning (CL) has emerged as a paramount area in Artificial Intelligence (AI) because of its ability to learn multiple tasks sequentially without significant performance degradation. Despite the growing interest ...
[ -0.10701829940080643, 0.009623408317565918, -0.030573980882763863, 0.002965189516544342, -0.029899967834353447, -0.0037655800115317106, 0.0059584625996649265, 0.012450519017875195, -0.018619609996676445, -0.0029113618656992912, -0.013414732180535793, -0.033157698810100555, -0.013096448034048...
wacv_2025_4718561761
4718561761
wacv
2,025
Towards Unsupervised Blind Face Restoration using Diffusion Prior
Blind face restoration methods have shown remarkable performance particularly when trained on large-scale synthetic datasets with supervised learning. These datasets are often generated by simulating low-quality face images with a handcrafted image degradation pipeline. The models trained on such synthetic degradations...
Tianshu Kuai; Sina Honari; Igor Gilitschenski; Alex Levinshtein
Samsung AI Center Toronto + University of Toronto; Samsung AI Center Toronto; University of Toronto + Vector Institute for AI; Samsung AI Center Toronto
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Kuai_Towards_Unsupervised_Blind_Face_Restoration_using_Diffusion_Prior_WACV_2025_paper.html
0
2410.04618
Towards Unsupervised Blind Face Restoration using Diffusion Prior Blind face restoration methods have shown remarkable performance particularly when trained on large-scale synthetic datasets with supervised learning. These datasets are often generated by simulating low-quality face images with a handcrafted image degra...
[ -0.04202786460518837, -0.009320254437625408, -0.022182388231158257, -0.0013978099450469017, -0.01561895664781332, -0.03457896411418915, 0.01744466833770275, 0.04315980523824692, 0.015454643405973911, 0.05265350267291069, -0.05006099492311478, -0.011711935512721539, -0.009804068133234978, 0...
wacv_2025_28166f1484
28166f1484
wacv
2,025
Towards Utilising a Range of Neural Activations for Comprehending Representational Associations
Recent efforts to understand intermediate representations in deep neural networks have commonly attempted to label individual neurons and combinations of neurons that make up linear directions in the latent space by examining extremal neuron activations and the highest direction projections. In this paper we show that ...
Laura O'Mahony; Nikola S. Nikolov; David JP O'Sullivan
University of Limerick; University of Limerick; University of Limerick
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/OMahony_Towards_Utilising_a_Range_of_Neural_Activations_for_Comprehending_Representational_WACV_2025_paper.html
1
2411.10019
Towards Utilising a Range of Neural Activations for Comprehending Representational Associations Recent efforts to understand intermediate representations in deep neural networks have commonly attempted to label individual neurons and combinations of neurons that make up linear directions in the latent space by examinin...
[ -0.0012153390562161803, -0.04906006157398224, -0.02201506495475769, 0.013759415596723557, -0.026552937924861908, -0.023363670334219933, 0.04326470568776131, -0.008966400288045406, 0.006214517168700695, 0.01629260554909706, -0.007913941517472267, -0.016602421179413795, -0.045451629906892776, ...
wacv_2025_b773fc6994
b773fc6994
wacv
2,025
Towards Zero-Shot 3D Anomaly Localization
3D anomaly detection and localization is of great significance for industrial inspection. Prior 3D anomaly detection and localization methods focus on the setting that the testing data share the same category as the training data which is normal. However in real-world applications the normal training data for the targe...
Yizhou Wang; Kuan-Chuan Peng; Yun Fu
Northeastern University+Mitsubishi Electric Research Laboratories; Mitsubishi Electric Research Laboratories; Northeastern University
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Wang_Towards_Zero-Shot_3D_Anomaly_Localization_WACV_2025_paper.html
1
2412.04304
Towards Zero-Shot 3D Anomaly Localization 3D anomaly detection and localization is of great significance for industrial inspection. Prior 3D anomaly detection and localization methods focus on the setting that the testing data share the same category as the training data which is normal. However in real-world applicati...
[ -0.0462028905749321, -0.014040420763194561, -0.031653422862291336, 0.0029756068252027035, 0.0027419081889092922, 0.026822110638022423, 0.007496863137930632, 0.013818291947245598, 0.01816832274198532, 0.007228456903249025, -0.02913595736026764, -0.007441330701112747, 0.009616346098482609, -...
wacv_2025_80712eee94
80712eee94
wacv
2,025
Towards a Training Free Approach for 3D Scene Editing
Text driven diffusion models have shown remarkable capabilities in editing images. However when editing 3D scenes existing works mostly rely on training a NeRF for 3D editing. Recent NeRF editing methods leverages edit operations by deploying 2D diffusion models and project these edits into 3D space. They require stron...
Vivek Madhavaram; Shivangana Rawat; Chaitanya Devaguptapu; Charu Sharma; Manohar Kaul
Machine Learning Lab, IIIT Hyderabad, India; Fujitsu Research India; Fujitsu Research India; Machine Learning Lab, IIIT Hyderabad, India; Fujitsu Research India
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Madhavaram_Towards_a_Training_Free_Approach_for_3D_Scene_Editing_WACV_2025_paper.html
0
2412.12766
Towards a Training Free Approach for 3D Scene Editing Text driven diffusion models have shown remarkable capabilities in editing images. However when editing 3D scenes existing works mostly rely on training a NeRF for 3D editing. Recent NeRF editing methods leverages edit operations by deploying 2D diffusion models and...
[ -0.015369626693427563, -0.03507724031805992, -0.04352208971977234, -0.011387213133275509, 0.03833073005080223, 0.004769118037074804, 0.005618047434836626, 0.028001343831419945, 0.04711337387561798, 0.01944982260465622, -0.011911682784557343, -0.040001921355724335, 0.0000491690261696931, 0....
wacv_2025_53ff3eda90
53ff3eda90
wacv
2,025
TrackDiffusion: Tracklet-Conditioned Video Generation via Diffusion Models
Despite remarkable achievements in video synthesis achieving granular control over complex dynamics such as nuanced movement among multiple interacting objects still presents a significant hurdle for dynamic world modeling compounded by the necessity to manage appearance and disappearance drastic scale changes and ensu...
Pengxiang Li; Kai Chen; Zhili Liu; Ruiyuan Gao; Lanqing Hong; Dit-Yan Yeung; Huchuan Lu; Xu Jia
Dalian University of Technology; Hong Kong University of Science and Technology; Huawei Noah’s Ark Lab + The Chinese University of Hong Kong; The Chinese University of Hong Kong; Huawei Noah’s Ark Lab; Hong Kong University of Science and Technology; Dalian University of Technology; Dalian University of Technology
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Li_TrackDiffusion_Tracklet-Conditioned_Video_Generation_via_Diffusion_Models_WACV_2025_paper.html
5
2312.00651
TrackDiffusion: Tracklet-Conditioned Video Generation via Diffusion Models Despite remarkable achievements in video synthesis achieving granular control over complex dynamics such as nuanced movement among multiple interacting objects still presents a significant hurdle for dynamic world modeling compounded by the nece...
[ -0.057887494564056396, -0.048962268978357315, -0.0508773997426033, 0.03743535652756691, -0.02177104540169239, -0.004647802095860243, -0.03503241389989853, 0.011445607990026474, 0.025998784229159355, -0.0035502160899341106, -0.04021771624684334, -0.03588157519698143, 0.01085842214524746, 0....
wacv_2025_a893b3dd71
a893b3dd71
wacv
2,025
Training-Free Medical Image Inverses via Bi-Level Guided Diffusion Models
In medical imaging inverse problems aim to infer high-fidelity images from incomplete noisy measurements minimizing expenses and risks to patients in clinical settings. Diffusion models have recently emerged as a promising solution to such practical challenges proving particularly useful for the training-free inference...
Hossein Askari; Fred Roosta; Hongfu Sun
The University of Queensland, Brisbane, Australia; The University of Queensland, Brisbane, Australia; The University of Queensland, Brisbane, Australia
Poster
main
https://github.com/hosseinaskari-cs/BGDM
https://openaccess.thecvf.com/content/WACV2025/html/Askari_Training-Free_Medical_Image_Inverses_via_Bi-Level_Guided_Diffusion_Models_WACV_2025_paper.html
1
Training-Free Medical Image Inverses via Bi-Level Guided Diffusion Models In medical imaging inverse problems aim to infer high-fidelity images from incomplete noisy measurements minimizing expenses and risks to patients in clinical settings. Diffusion models have recently emerged as a promising solution to such practi...
[ -0.10181944072246552, -0.04359510540962219, -0.03607933595776558, -0.0013006299268454313, 0.011118214577436447, 0.0004840217879973352, 0.011584521271288395, 0.010185601189732552, 0.022035276517271996, 0.030885960906744003, -0.07062260806560516, 0.0017909377347677946, 0.02918531373143196, 0...
wacv_2025_6b251effa4
6b251effa4
wacv
2,025
Transferable-Guided Attention is All You Need for Video Domain Adaptation
Unsupervised domain adaptation (UDA) in videos is a challenging task that remains not well explored compared to image-based UDA techniques. Although vision transformers (ViT) achieve state-of-the-art performance in many computer vision tasks their use in video UDA has been little explored. Our key idea is to use transf...
André Sacilotti; Samuel Felipe dos Santos; Nicu Sebe; Jurandy Almeida
University of S ˜ao Paulo; Federal University of S ˜ao Carlos; University of Trento; Federal University of S ˜ao Carlos
Poster
main
https://github.com/Andre-Sacilotti/transferattn-project-code
https://openaccess.thecvf.com/content/WACV2025/html/Sacilotti_Transferable-Guided_Attention_is_All_You_Need_for_Video_Domain_Adaptation_WACV_2025_paper.html
1
2407.01375
Transferable-Guided Attention is All You Need for Video Domain Adaptation Unsupervised domain adaptation (UDA) in videos is a challenging task that remains not well explored compared to image-based UDA techniques. Although vision transformers (ViT) achieve state-of-the-art performance in many computer vision tasks thei...
[ -0.02748224325478077, -0.017605531960725784, -0.06553319841623306, 0.017075298354029655, -0.024139080196619034, 0.018585113808512688, 0.007274975534528494, -0.010182269848883152, 0.004969810601323843, 0.028974084183573723, -0.012114474549889565, -0.009939621202647686, -0.029980629682540894, ...
wacv_2025_1135558fee
1135558fee
wacv
2,025
Transferring Foundation Models for Generalizable Robotic Manipulation
Improving the generalization capabilities of general-purpose robotic manipulation in real world has long been a significant challenge. Existing approaches often rely on collecting large-scale robotic data which is costly and time-consuming. However due to insufficient diversity of data they typically suffer from limiti...
Jiange Yang; Wenhui Tan; Chuhao Jin; Keling Yao; Bei Liu; Jianlong Fu; Ruihua Song; Gangshan Wu; Limin Wang
State Key Laboratory for Novel Software Technology, Nanjing University, China; Renmin University of China; Renmin University of China; The Chinese University of Hong Kong, Shenzhen; Microsoft Research; Microsoft Research; Renmin University of China; State Key Laboratory for Novel Software Technology, Nanjing University...
Poster
main
https://github.com/MCG-NJU/TPM
https://openaccess.thecvf.com/content/WACV2025/html/Yang_Transferring_Foundation_Models_for_Generalizable_Robotic_Manipulation_WACV_2025_paper.html
8
2306.05716
Transferring Foundation Models for Generalizable Robotic Manipulation Improving the generalization capabilities of general-purpose robotic manipulation in real world has long been a significant challenge. Existing approaches often rely on collecting large-scale robotic data which is costly and time-consuming. However d...
[ 0.00275396928191185, -0.003800386330112815, -0.0012721149250864983, 0.02068212628364563, -0.009948941878974438, -0.026481876149773598, -0.008239110000431538, -0.013933990150690079, 0.019788453355431557, 0.025606442242860794, -0.031314998865127563, 0.0062009901739656925, 0.0018682762747630477...
wacv_2025_79d9a6c006
79d9a6c006
wacv
2,025
Transientangelo: Few-Viewpoint Surface Reconstruction using Single-Photon Lidar
We consider the problem of few-viewpoint 3D surface reconstruction using raw measurements from a lidar system. Lidar captures 3D scene geometry by emitting pulses of light to a target and recording the speed-of-light time delay of the reflected light. However conventional lidar systems do not output the raw captured wa...
Weihan Luo; Anagh Malik; David B Lindell
University of Toronto; University of Toronto + Vector Institute; University of Toronto + Vector Institute
Poster
main
https://weihan1.github.io/transientangelo/
https://openaccess.thecvf.com/content/WACV2025/html/Luo_Transientangelo_Few-Viewpoint_Surface_Reconstruction_using_Single-Photon_Lidar_WACV_2025_paper.html
1
2408.12191
Transientangelo: Few-Viewpoint Surface Reconstruction using Single-Photon Lidar We consider the problem of few-viewpoint 3D surface reconstruction using raw measurements from a lidar system. Lidar captures 3D scene geometry by emitting pulses of light to a target and recording the speed-of-light time delay of the refle...
[ -0.05514761805534363, -0.005640511400997639, -0.054673779755830765, -0.019409190863370895, -0.015135524794459343, 0.008328638970851898, -0.008920938707888126, 0.06298419088125229, -0.005804532673209906, 0.009221644140779972, -0.038198746740818024, -0.015645813196897507, 0.014652572572231293,...
wacv_2025_db8437132a
db8437132a
wacv
2,025
Treading Towards Privacy-Preserving Table Structure Recognition
We present TabGuard a privacy-preserving framework for an end-to-end secure Table Structure Recognition. TabGuard masks all the contents of the table locally and utilizes the masked table image for structure recognition. Our method is simple yet effective for detecting table cells while preserving the inherent table al...
Sachin Raja; Ajoy Mondal; C.V. Jawahar
IIIT Hyderabad; IIIT Hyderabad; IIIT Hyderabad
Poster
main
https://github.com/sachinraja13/TabGuard
https://openaccess.thecvf.com/content/WACV2025/html/Raja_Treading_Towards_Privacy-Preserving_Table_Structure_Recognition_WACV_2025_paper.html
0
Treading Towards Privacy-Preserving Table Structure Recognition We present TabGuard a privacy-preserving framework for an end-to-end secure Table Structure Recognition. TabGuard masks all the contents of the table locally and utilizes the masked table image for structure recognition. Our method is simple yet effective ...
[ -0.0036912267096340656, 0.0032756426371634007, -0.03479699045419693, 0.0028740670531988144, -0.03619783744215965, -0.055249329656362534, 0.05188729986548424, 0.006747403647750616, 0.016931546851992607, 0.04815171658992767, -0.02028423734009266, -0.002526190597563982, -0.03268638625741005, ...
wacv_2025_8253ed54d8
8253ed54d8
wacv
2,025
TreeFormer: Single-View Plant Skeleton Estimation via Tree-Constrained Graph Generation
Accurate estimation of plant skeletal structure (e.g. branching structure) from images is essential for smart agriculture and plant science. Unlike human skeletons with fixed topology plant skeleton estimation presents a unique challenge i.e. estimating arbitrary tree graphs from images. While recent graph generation m...
Xinpeng Liu; Hiroaki Santo; Yosuke Toda; Fumio Okura
Osaka University; Osaka University; Phytometrics+Nagoya University; Osaka University
Poster
main
https://github.com/huntorochi/TreeFormer/
https://openaccess.thecvf.com/content/WACV2025/html/Liu_TreeFormer_Single-View_Plant_Skeleton_Estimation_via_Tree-Constrained_Graph_Generation_WACV_2025_paper.html
0
2411.16132
TreeFormer: Single-View Plant Skeleton Estimation via Tree-Constrained Graph Generation Accurate estimation of plant skeletal structure (e.g. branching structure) from images is essential for smart agriculture and plant science. Unlike human skeletons with fixed topology plant skeleton estimation presents a unique chal...
[ -0.033327169716358185, -0.005722985137254, -0.05277712270617485, -0.006651774980127811, -0.002497261157259345, -0.07554157823324203, -0.019686704501509666, 0.011682720854878426, 0.022345593199133873, 0.027044177055358887, -0.008545777760446072, -0.013631357811391354, -0.02043337933719158, ...
wacv_2025_74ce9ef5a3
74ce9ef5a3
wacv
2,025
Tumor Synthesis Conditioned on Radiomics
Due to privacy concerns obtaining large datasets is challenging in medical image analysis especially with 3D modalities like Computed Tomography (CT) and Magnetic Resonance Imaging (MRI). Existing generative models developed to address this issue often face limitations in output diversity and thus cannot accurately rep...
Jonghun Kim; Inye Na; Eun Sook Ko; Hyunjin Park
Department of Electrical and Computer Engineering, Sungkyunkwan University, Suwon, Korea; Department of Electrical and Computer Engineering, Sungkyunkwan University, Suwon, Korea; Department of Radiology and Center for Imaging Science, Samsung Medical Center, Sungkyunkwan University School of Medicine, Suwon, Korea; De...
Poster
main
github.com/jongdory/TS-Radiomics
https://openaccess.thecvf.com/content/WACV2025/html/Kim_Tumor_Synthesis_Conditioned_on_Radiomics_WACV_2025_paper.html
0
Tumor Synthesis Conditioned on Radiomics Due to privacy concerns obtaining large datasets is challenging in medical image analysis especially with 3D modalities like Computed Tomography (CT) and Magnetic Resonance Imaging (MRI). Existing generative models developed to address this issue often face limitations in output...
[ 0.00482176011428237, 0.012377978302538395, -0.050241101533174515, 0.00504051661118865, -0.0032107089646160603, -0.003707468742504716, 0.014365017414093018, 0.047105588018894196, 0.0031674134079366922, 0.01173082273453474, -0.02541222609579563, 0.019578715786337852, -0.009944310411810875, 0...
wacv_2025_91655d1ee8
91655d1ee8
wacv
2,025
Tuned Contrastive Learning
In recent times contrastive learning based loss functions have become increasingly popular for visual self-supervised representation learning owing to their state-of-the-art (SOTA) performance. Most of the modern contrastive learning methods generalize only to one positive and multiple negatives per anchor in a batch. ...
Chaitanya Animesh; Manmohan Chandraker
UC San Diego+Otter.ai, Inc.; UC San Diego
Poster
main
https://github.com/chaitanyaanimesh/Tuned-Contrastive-Learning
https://openaccess.thecvf.com/content/WACV2025/html/Animesh_Tuned_Contrastive_Learning_WACV_2025_paper.html
0
2305.10675
Tuned Contrastive Learning In recent times contrastive learning based loss functions have become increasingly popular for visual self-supervised representation learning owing to their state-of-the-art (SOTA) performance. Most of the modern contrastive learning methods generalize only to one positive and multiple negati...
[ -0.06896480917930603, -0.01901262253522873, -0.029492752626538277, -0.028899187222123146, 0.02695155330002308, 0.004382177721709013, -0.009654702618718147, 0.028212878853082657, -0.008439749479293823, 0.011351927183568478, -0.007679244969040155, -0.02802738919854164, -0.018150098621845245, ...
wacv_2025_dbc6474478
dbc6474478
wacv
2,025
U-MixFormer: UNet-Like Transformer with Mix-Attention for Efficient Semantic Segmentation
Semantic segmentation has witnessed remarkable advancements with the adaptation of the Transformer architecture. Parallel to the strides made by the Transformer CNN-based U-Net has seen significant progress especially in high-resolution medical imaging and remote sensing. This dual success inspired us to merge both str...
Seul-Ki Yeom; Julian von Klitzing
Nota AI GmbH; Nota AI GmbH
Poster
main
https://github.com/julian-klitzing/u-mixformer
https://openaccess.thecvf.com/content/WACV2025/html/Yeom_U-MixFormer_UNet-Like_Transformer_with_Mix-Attention_for_Efficient_Semantic_Segmentation_WACV_2025_paper.html
14
U-MixFormer: UNet-Like Transformer with Mix-Attention for Efficient Semantic Segmentation Semantic segmentation has witnessed remarkable advancements with the adaptation of the Transformer architecture. Parallel to the strides made by the Transformer CNN-based U-Net has seen significant progress especially in high-reso...
[ -0.01568283699452877, -0.0262291356921196, -0.06065487861633301, -0.01504532340914011, 0.017586272209882736, 0.008743044920265675, -0.0018624508520588279, 0.02349693514406681, 0.019908642396330833, 0.01872468926012516, -0.015090860426425934, -0.024699103087186813, 0.010345936752855778, 0.0...
wacv_2025_0c2a5f8b22
0c2a5f8b22
wacv
2,025
UAL-Bench: The First Comprehensive Unusual Activity Localization Benchmark
Localizing unusual activities in videos such as abnormal behaviors or traffic incidents holds practical significance. However pretrained foundation models struggle with localizing diverse unusual events likely because of their insufficient representation in the models' pretraining datasets. To explore foundation models...
Hasnat Md Abdullah; Tian Liu; Kangda Wei; Shu Kong; Ruihong Huang
Texas A&M University; Texas A&M University; Texas A&M University; University of Macau+Institute of Collaborative Innovation; Texas A&M University
Poster
main
https://github.com/Hasnat79/UAL_Bench
https://openaccess.thecvf.com/content/WACV2025/html/Abdullah_UAL-Bench_The_First_Comprehensive_Unusual_Activity_Localization_Benchmark_WACV_2025_paper.html
3
UAL-Bench: The First Comprehensive Unusual Activity Localization Benchmark Localizing unusual activities in videos such as abnormal behaviors or traffic incidents holds practical significance. However pretrained foundation models struggle with localizing diverse unusual events likely because of their insufficient repre...
[ -0.029437482357025146, -0.042691670358181, -0.049904584884643555, 0.006526406854391098, -0.04704871028661728, 0.03771219775080681, -0.007501248735934496, 0.002972581423819065, -0.010709531605243683, 0.010370854288339615, -0.013968158513307571, -0.01260429434478283, 0.0033089707139879465, 0...
wacv_2025_50eaaea4e4
50eaaea4e4
wacv
2,025
UCDR-Adapter: Exploring Adaptation of Pre-Trained Vision-Language Models for Universal Cross-Domain Retrieval
Universal Cross-Domain Retrieval (UCDR) retrieves relevant images from unseen domains and classes without semantic labels ensuring robust generalization. Existing methods commonly employ prompt tuning with pre-trained vision-language models but are inherently limited by static prompts reducing adaptability. We propose ...
Haoyu Jiang; Zhi-Qi Cheng; Gabriel Moreira; Jiawen Zhu; Jingdong Sun; Bukun Ren; Jun-Yan He; Qi Dai; Xian-Sheng Hua
Zhejiang University; Carnegie Mellon University; Dalian University of Technology; DAMO Academy, Alibaba Group; Microsoft Research; Zhejiang University; Carnegie Mellon University; DAMO Academy, Alibaba Group; Microsoft Research
Poster
main
https://github.com/fine68/UCDR2024
https://openaccess.thecvf.com/content/WACV2025/html/Jiang_UCDR-Adapter_Exploring_Adaptation_of_Pre-Trained_Vision-Language_Models_for_Universal_Cross-Domain_WACV_2025_paper.html
0
UCDR-Adapter: Exploring Adaptation of Pre-Trained Vision-Language Models for Universal Cross-Domain Retrieval Universal Cross-Domain Retrieval (UCDR) retrieves relevant images from unseen domains and classes without semantic labels ensuring robust generalization. Existing methods commonly employ prompt tuning with pre-...
[ -0.05896998941898346, -0.012383697554469109, -0.06285293400287628, -0.005438847467303276, -0.02376944199204445, -0.02093888260424137, -0.01084140595048666, -0.01961432583630085, 0.02763424441218376, 0.00987974088639021, -0.025620192289352417, -0.030410369858145714, -0.014978377148509026, 0...
wacv_2025_373df0d0cd
373df0d0cd
wacv
2,025
USWformer: Efficient Sparse Wavelet Transformer for Underwater Image Enhancement
Transformer-based methods have shown great promise in underwater image enhancement (UIE) tasks due to their capability to model long-range dependencies which are vital for reconstructing clear images. While numerous effective attention mechanisms have been devised to handle the computational requirements of transformer...
Priyanka Mishra; Nancy Mehta; Santosh Kumar Vipparthi; Subrahmanyam Murala
Indian Institute of Technology Ropar, INDIA; University of Würzburg, Germany; Indian Institute of Technology Ropar, INDIA; Trinity College Dublin, Ireland
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Mishra_USWformer_Efficient_Sparse_Wavelet_Transformer_for_Underwater_Image_Enhancement_WACV_2025_paper.html
0
USWformer: Efficient Sparse Wavelet Transformer for Underwater Image Enhancement Transformer-based methods have shown great promise in underwater image enhancement (UIE) tasks due to their capability to model long-range dependencies which are vital for reconstructing clear images. While numerous effective attention mec...
[ -0.05284753441810608, 0.015705730766057968, -0.04089174419641495, -0.013028512708842754, 0.01650339365005493, -0.01588910073041916, 0.04686963930726051, 0.010653858073055744, 0.004717220086604357, 0.04184527322649956, -0.011845769360661507, 0.01025961060076952, -0.02220623381435871, 0.0086...
wacv_2025_1f73faa844
1f73faa844
wacv
2,025
UW-GS: Distractor-Aware 3D Gaussian Splatting for Enhanced Underwater Scene Reconstruction
3D Gaussian splatting (3DGS) offers the capability to achieve real-time high quality 3D scene rendering. However 3DGS assumes that the scene is in a clear medium environment and struggles to generate satisfactory representations in underwater scenes where light absorption and scattering are prevalent and moving objects...
Haoran Wang; Nantheera Anantrasirichai; Fan Zhang; David Bull
School of Computer Science, University of Bristol, Bristol, UK; School of Computer Science, University of Bristol, Bristol, UK; School of Computer Science, University of Bristol, Bristol, UK; School of Computer Science, University of Bristol, Bristol, UK
Poster
main
https://github.com/WangHaoran16/UW-GS
https://openaccess.thecvf.com/content/WACV2025/html/Wang_UW-GS_Distractor-Aware_3D_Gaussian_Splatting_for_Enhanced_Underwater_Scene_Reconstruction_WACV_2025_paper.html
7
UW-GS: Distractor-Aware 3D Gaussian Splatting for Enhanced Underwater Scene Reconstruction 3D Gaussian splatting (3DGS) offers the capability to achieve real-time high quality 3D scene rendering. However 3DGS assumes that the scene is in a clear medium environment and struggles to generate satisfactory representations ...
[ -0.07394317537546158, -0.0264270082116127, -0.06218119338154793, -0.04386396333575249, -0.017174744978547096, -0.02552800253033638, -0.01760551892220974, -0.00990778673440218, -0.01994667761027813, 0.04071744531393051, -0.01059140544384718, 0.058173127472400665, 0.007664956152439117, 0.007...
wacv_2025_4e5dbd8e55
4e5dbd8e55
wacv
2,025
UnDIVE: Generalized Underwater Video Enhancement using Generative Priors
With the rise of marine exploration underwater imaging has gained significant attention as a research topic. Underwater video enhancement has become crucial for real-time computer vision tasks in marine exploration. However most existing methods focus on enhancing individual frames and neglect video temporal dynamics l...
Suhas Srinath; Aditya Chandrasekar; Hemang Jamadagni; Rajiv Soundararajan; Prathosh A P
Indian Institute of Science; Indian Institute of Science + Qualcomm; National Institute of Technology Karnataka; Indian Institute of Science; Indian Institute of Science
Poster
main
github.com/suhas-srinath/undive
https://openaccess.thecvf.com/content/WACV2025/html/Srinath_UnDIVE_Generalized_Underwater_Video_Enhancement_using_Generative_Priors_WACV_2025_paper.html
1
2411.05886
UnDIVE: Generalized Underwater Video Enhancement using Generative Priors With the rise of marine exploration underwater imaging has gained significant attention as a research topic. Underwater video enhancement has become crucial for real-time computer vision tasks in marine exploration. However most existing methods f...
[ -0.08129387348890305, -0.015915410593152046, -0.04383930191397667, -0.0014917796943336725, -0.010542219504714012, 0.02680099569261074, 0.007832423783838749, 0.04313401132822037, -0.03125545382499695, 0.024425283074378967, -0.0067466492764651775, -0.04168631136417389, -0.020007945597171783, ...
wacv_2025_f6008a510f
f6008a510f
wacv
2,025
Uncertainty Aware Interest Point Detection and Description
Interest point detection and description play an important role in many visual tasks including image registration pose estimation 3D reconstruction and more. State-of-the-art interest point detection techniques are based on deep neural networks (NNs) which are prone to produce overconfident predictions. However calibra...
Jingbo Zeng; Zaiwang Gu; Weide Liu; Lile Cai; Jun Cheng
School of Electrical and Electronic Engineering, Nanyang Technological University + Institute for Infocomm Research (I2R), A*STAR, Singapore; Institute for Infocomm Research (I2R), A*STAR, Singapore; Boston Children’s Hospital and Harvard Medical School, Boston, MA; Institute for Infocomm Research (I2R), A*STAR, Singap...
Poster
main
https://github.com/JingboZeng/UAPoint
https://openaccess.thecvf.com/content/WACV2025/html/Zeng_Uncertainty_Aware_Interest_Point_Detection_and_Description_WACV_2025_paper.html
0
Uncertainty Aware Interest Point Detection and Description Interest point detection and description play an important role in many visual tasks including image registration pose estimation 3D reconstruction and more. State-of-the-art interest point detection techniques are based on deep neural networks (NNs) which are ...
[ 0.003965464420616627, -0.040362272411584854, -0.03556511923670769, 0.014400656335055828, -0.004990144167095423, -0.012470765970647335, 0.035785675048828125, 0.008215817622840405, 0.0000011465004945421242, 0.00896939355880022, -0.04396473616361618, -0.015328841283917427, -0.001439376384951174...
wacv_2025_f5b1b73630
f5b1b73630
wacv
2,025
Uncertainty Awareness Enables Efficient Labeling for Cancer Subtyping in Digital Pathology
Machine-learning-assisted cancer subtyping is a promising avenue in digital pathology. Cancer subtyping models however require careful training using expert annotations so that they can be inferred with a degree of known certainty (or uncertainty). To this end we introduce the concept of uncertainty awareness into a se...
Nirhoshan Sivaroopan; Chamuditha Jayanga Galappaththige; Chalani Ekanayake; Hasindri Watawana; Ranga Rodrigo; Chamira U.S. Edussooriya; Dushan N. Wadduwage
University of Moratuwa; University of Moratuwa; University of Moratuwa; University of Moratuwa; University of Moratuwa; University of Moratuwa; Harvard University + Old Dominion University
Poster
main
https://github.com/Nirhoshan/AI-for-histopathology
https://openaccess.thecvf.com/content/WACV2025/html/Sivaroopan_Uncertainty_Awareness_Enables_Efficient_Labeling_for_Cancer_Subtyping_in_Digital_WACV_2025_paper.html
0
Uncertainty Awareness Enables Efficient Labeling for Cancer Subtyping in Digital Pathology Machine-learning-assisted cancer subtyping is a promising avenue in digital pathology. Cancer subtyping models however require careful training using expert annotations so that they can be inferred with a degree of known certaint...
[ -0.0838458463549614, -0.04825712740421295, -0.027313973754644394, 0.036833591759204865, -0.017208535224199295, -0.03233008086681366, 0.03308066353201866, 0.04463235288858414, -0.019203994423151016, 0.050014596432447433, -0.022407710552215576, -0.026453547179698944, -0.004622505512088537, 0...
wacv_2025_835c3abb20
835c3abb20
wacv
2,025
Uncertainty and Energy Based Loss Guided Semi-Supervised Semantic Segmentation
Semi-supervised (SS) semantic segmentation exploits both labeled and unlabeled images to overcome tedious and costly pixel-level annotation problems. Pseudolabel supervision is one of the core approaches of training networks with both pseudo labels and ground-truth labels. This work uses aleatoric or data uncertainty a...
Rini Smita Thakur; Vinod K Kurmi
Indian Institute of Science Education and Research Bhopal, India; Indian Institute of Science Education and Research Bhopal, India
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Thakur_Uncertainty_and_Energy_Based_Loss_Guided_Semi-Supervised_Semantic_Segmentation_WACV_2025_paper.html
0
2501.01640
Uncertainty and Energy Based Loss Guided Semi-Supervised Semantic Segmentation Semi-supervised (SS) semantic segmentation exploits both labeled and unlabeled images to overcome tedious and costly pixel-level annotation problems. Pseudolabel supervision is one of the core approaches of training networks with both pseudo...
[ -0.07062960416078568, -0.008888603188097477, -0.04663175716996193, 0.024329612031579018, -0.0037277759984135628, -0.045452140271663666, 0.04386702924966812, 0.017261123284697533, -0.010671853087842464, 0.037858352065086365, -0.01580503210425377, -0.001734867342747748, -0.0009066006168723106,...
wacv_2025_05697de2d2
05697de2d2
wacv
2,025
Uncertainty-Aware Online Extrinsic Calibration: A Conformal Prediction Approach
Accurate sensor calibration is crucial for autonomous systems yet its uncertainty quantification remains underexplored. We present the first approach to integrate uncertainty awareness into online extrinsic calibration combining Monte Carlo Dropout with Conformal Prediction to generate prediction intervals with a guara...
Mathieu Cocheteux; Julien Moreau; Franck Davoine
Universit ´e de technologie de Compi `egne, CNRS, Heudiasyc, France; Universit ´e de technologie de Compi `egne, CNRS, Heudiasyc, France; CNRS, INSA Lyon, UCBL, LIRIS, UMR5205, France
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Cocheteux_Uncertainty-Aware_Online_Extrinsic_Calibration_A_Conformal_Prediction_Approach_WACV_2025_paper.html
2
2501.06878
Uncertainty-Aware Online Extrinsic Calibration: A Conformal Prediction Approach Accurate sensor calibration is crucial for autonomous systems yet its uncertainty quantification remains underexplored. We present the first approach to integrate uncertainty awareness into online extrinsic calibration combining Monte Carlo...
[ -0.05591373145580292, -0.0024078702554106712, -0.034946080297231674, 0.04030637443065643, -0.038753215223550797, -0.027540164068341255, -0.0003746759903151542, -0.0036106216721236706, -0.00448900880292058, 0.05030720308423042, -0.024149730801582336, -0.008139880374073982, -0.0109668187797069...
wacv_2025_3efc19f5f7
3efc19f5f7
wacv
2,025
Uncertainty-Aware Regularization for Image-to-Image Translation
The importance of quantifying uncertainty in deep networks has become paramount for reliable real-world applications. In this paper we propose a method to improve uncertainty estimation in medical Image-to-Image (I2I) translation. Our model integrates aleatoric uncertainty and employs Uncertainty-Aware Regularization (...
Anuja Vats; Ivar Farup; Marius Pedersen; Kiran Raja
;;;
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Vats_Uncertainty-Aware_Regularization_for_Image-to-Image_Translation_WACV_2025_paper.html
0
2412.01705
Uncertainty-Aware Regularization for Image-to-Image Translation The importance of quantifying uncertainty in deep networks has become paramount for reliable real-world applications. In this paper we propose a method to improve uncertainty estimation in medical Image-to-Image (I2I) translation. Our model integrates alea...
[ -0.058951422572135925, 0.003112097503617406, -0.021226219832897186, 0.01911286637187004, -0.003628849284723401, 0.015738917514681816, 0.005338996648788452, 0.02684328891336918, -0.00008964949665823951, 0.013699717819690704, -0.034722018986940384, 0.001802838291041553, -0.015145695768296719, ...
wacv_2025_40ecb8ab39
40ecb8ab39
wacv
2,025
Uncertainty-Based Data-Wise Label Smoothing for Calibrating Multiple Instance Learning in Histopathology Image Classification
Deep neural networks (DNNs) have transformed biomedical image analysis particularly in histopathology with Whole Slide Images (WSIs) classification. However training DNNs requires large annotated datasets which is challenging due to the high heterogeneity and high resolution of WSIs. Multiple Instance Learning (MIL) ha...
Hyeongmin Park; Sungrae Hong; Chanjae Song; Jongwoo Kim; Mun Yong Yi
Korea Advanced Institute of Science and Technology; Korea Advanced Institute of Science and Technology; Korea Advanced Institute of Science and Technology; Korea Advanced Institute of Science and Technology; Korea Advanced Institute of Science and Technology
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Park_Uncertainty-Based_Data-Wise_Label_Smoothing_for_Calibrating_Multiple_Instance_Learning_in_WACV_2025_paper.html
0
Uncertainty-Based Data-Wise Label Smoothing for Calibrating Multiple Instance Learning in Histopathology Image Classification Deep neural networks (DNNs) have transformed biomedical image analysis particularly in histopathology with Whole Slide Images (WSIs) classification. However training DNNs requires large annotate...
[ -0.09201689809560776, -0.019941914826631546, -0.03966245427727699, 0.008255325257778168, -0.01996036246418953, -0.017608284950256348, 0.02082740142941475, 0.0025734477676451206, 0.02029241994023323, 0.01593877375125885, -0.02269061468541622, 0.004289079457521439, 0.01283956691622734, 0.043...
wacv_2025_a316f01c68
a316f01c68
wacv
2,025
Uncertainty-Guided Cross Attention Ensemble Mean Teacher for Semi-Supervised Medical Image Segmentation
This work proposes a novel framework Uncertainty-Guided Cross Attention Ensemble Mean Teacher (UG-CEMT) for achieving state-of-the-art performance in semi-supervised medical image segmentation. UG-CEMT leverages the strengths of co-training and knowledge distillation by combining a Cross-attention Ensemble Mean Teacher...
Meghana Karri; Amit Soni Arya; Koushik Biswas; Nicolo Gennaro; Vedat Cicek; Gorkem Durak; Yury S. Velichko; Ulas Bagci
Department of Radiology, Northwestern University, Chicago, IL, USA; School of Computer Science Engineering and Technology, Bennett University, Greater Noida, UP, India; Department of Radiology, Northwestern University, Chicago, IL, USA; Department of Radiology, Northwestern University, Chicago, IL, USA; Department of R...
Poster
main
https://github.com/Meghnak13/UG-CEMT
https://openaccess.thecvf.com/content/WACV2025/html/Karri_Uncertainty-Guided_Cross_Attention_Ensemble_Mean_Teacher_for_Semi-Supervised_Medical_Image_WACV_2025_paper.html
0
2412.15380
Uncertainty-Guided Cross Attention Ensemble Mean Teacher for Semi-Supervised Medical Image Segmentation This work proposes a novel framework Uncertainty-Guided Cross Attention Ensemble Mean Teacher (UG-CEMT) for achieving state-of-the-art performance in semi-supervised medical image segmentation. UG-CEMT leverages the ...
[ -0.0700421929359436, -0.03909331560134888, -0.062971331179142, 0.005178198218345642, 0.028579583391547203, -0.03317008540034294, 0.02000940963625908, 0.025303296744823456, -0.017778944224119186, 0.010560007765889168, -0.016187075525522232, -0.001798949670046568, 0.005400319583714008, 0.032...
wacv_2025_515a442b6b
515a442b6b
wacv
2,025
Uncertainty-Guided Metric Learning without Labels
Unsupervised metric learning aims to learn the discriminative representations by grouping similar examples in the absence of labels. Many unsupervised metric learning algorithms combine clustering-based pseudo-label generation with embedding fine-tuning. However pseudo-labels can be unreliable and noisy. This could aff...
Dhanunjaya Varma Devalraju; C Chandra Sekhar
Department of Computer Science and Engineering, Indian Institute of Technology Madras, India; Department of Computer Science and Engineering, Indian Institute of Technology Madras, India
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Devalraju_Uncertainty-Guided_Metric_Learning_without_Labels_WACV_2025_paper.html
0
Uncertainty-Guided Metric Learning without Labels Unsupervised metric learning aims to learn the discriminative representations by grouping similar examples in the absence of labels. Many unsupervised metric learning algorithms combine clustering-based pseudo-label generation with embedding fine-tuning. However pseudo-...
[ -0.07997941970825195, -0.00792963057756424, -0.035798728466033936, -0.00022905001242179424, -0.0008169642533175647, -0.035318706184625626, 0.01780705153942108, 0.01775166392326355, -0.011114406399428844, 0.015093068592250347, -0.05202724039554596, -0.006000302731990814, 0.0025178194046020508...
wacv_2025_beed573bc1
beed573bc1
wacv
2,025
Unconstrained Open Vocabulary Image Classification: Zero-Shot Transfer from Text to Image via CLIP Inversion
We introduce NOVIC an innovative real-time uNconstrained Open Vocabulary Image Classifier that uses an autoregressive transformer to generatively output classification labels as language. Leveraging the extensive knowledge of CLIP models NOVIC harnesses the embedding space to enable zero-shot transfer from pure text to...
Philipp Allgeuer; Kyra Ahrens; Stefan Wermter
University of Hamburg; University of Hamburg; University of Hamburg
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Allgeuer_Unconstrained_Open_Vocabulary_Image_Classification_Zero-Shot_Transfer_from_Text_to_WACV_2025_paper.html
1
2407.11211
Unconstrained Open Vocabulary Image Classification: Zero-Shot Transfer from Text to Image via CLIP Inversion We introduce NOVIC an innovative real-time uNconstrained Open Vocabulary Image Classifier that uses an autoregressive transformer to generatively output classification labels as language. Leveraging the extensiv...
[ -0.09270567446947098, -0.022717662155628204, -0.01152398344129324, 0.0487750768661499, 0.03620512783527374, -0.025121551007032394, -0.025323403999209404, 0.0020059163216501474, 0.03816860914230347, -0.00097600620938465, -0.020845932886004448, -0.03934302553534508, -0.012303871102631092, 0....
wacv_2025_784dd60944
784dd60944
wacv
2,025
Uni-SLAM: Uncertainty-Aware Neural Implicit SLAM for Real-Time Dense Indoor Scene Reconstruction
Neural implicit fields have recently emerged as a powerful representation method for multi-view surface reconstruction due to their simplicity and state-of-the-art performance. However reconstructing thin structures of indoor scenes while ensuring real-time performance remains a challenge for dense visual SLAM systems....
Shaoxiang Wang; Yaxu Xie; Chun-Peng Chang; Christen Millerdurai; Alain Pagani; Didier Stricker
German Research Center for Artificial Intelligence; German Research Center for Artificial Intelligence; German Research Center for Artificial Intelligence; German Research Center for Artificial Intelligence; German Research Center for Artificial Intelligence; German Research Center for Artificial Intelligence
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Wang_Uni-SLAM_Uncertainty-Aware_Neural_Implicit_SLAM_for_Real-Time_Dense_Indoor_Scene_WACV_2025_paper.html
0
Uni-SLAM: Uncertainty-Aware Neural Implicit SLAM for Real-Time Dense Indoor Scene Reconstruction Neural implicit fields have recently emerged as a powerful representation method for multi-view surface reconstruction due to their simplicity and state-of-the-art performance. However reconstructing thin structures of indo...
[ -0.0503535270690918, -0.008143758401274681, -0.031159546226263046, -0.00011847315909108147, -0.007195382844656706, 0.008516502566635609, 0.018618348985910416, 0.002840406959876418, -0.014881467446684837, 0.026365770027041435, -0.03210320323705673, 0.01892031915485859, 0.0006558414315804839, ...
wacv_2025_2033b3278b
2033b3278b
wacv
2,025
UniTMGE: Uniform Text-Motion Generation and Editing Model via Diffusion
Current methods have shown promising results in applying diffusion models to motion generation given text input. However these methods are limited to unimodal inputs and outputs restricted to motion generation alone and lacking multimodal control capabilities. To address these issues we introduce TMMGE a text-motion mu...
Ruoyu Wang; Yangfan He; Tengjiao Sun; Xiang Li; Tianyu Shi
Tsinghua University; University of Minnesota - Twin Cities+Henan Runtai Digital Technology Group Co., Ltd.; University of Southampton; Li Auto Inc.; University of Toronto
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Wang_UniTMGE_Uniform_Text-Motion_Generation_and_Editing_Model_via_Diffusion_WACV_2025_paper.html
0
UniTMGE: Uniform Text-Motion Generation and Editing Model via Diffusion Current methods have shown promising results in applying diffusion models to motion generation given text input. However these methods are limited to unimodal inputs and outputs restricted to motion generation alone and lacking multimodal control c...
[ -0.06459914892911911, -0.012039264664053917, -0.011865709908306599, 0.010175827890634537, -0.00016413546109106392, -0.02139298804104328, -0.04223790764808655, 0.03085632622241974, 0.0053208437748253345, 0.008307823911309242, -0.02692849189043045, -0.036592788994312286, 0.0007124906405806541,...
wacv_2025_636fffbe99
636fffbe99
wacv
2,025
Unified Framework for Open-World Compositional Zero-Shot Learning
Open-World Compositional Zero-Shot Learning (OW-CZSL) addresses the challenge of recognizing novel compositions of known primitives and entities. Even though prior works utilize language knowledge for recognition such approaches exhibit limited interactions between language-image modalities. Our approach primarily focu...
Hirunima Jayasekara; Khoi Pham; Nirat Saini; Abhinav Shrivastava
University of Maryland; University of Maryland; University of Maryland; University of Maryland
Poster
main
https://github.com/hirunima/OWCZSL
https://openaccess.thecvf.com/content/WACV2025/html/Jayasekara_Unified_Framework_for_Open-World_Compositional_Zero-Shot_Learning_WACV_2025_paper.html
0
2412.04083
Unified Framework for Open-World Compositional Zero-Shot Learning Open-World Compositional Zero-Shot Learning (OW-CZSL) addresses the challenge of recognizing novel compositions of known primitives and entities. Even though prior works utilize language knowledge for recognition such approaches exhibit limited interacti...
[ -0.029067521914839745, 0.004622492473572493, -0.03472978621721268, 0.03187099099159241, 0.002318162703886628, -0.012025396339595318, 0.030100379139184952, 0.005537768825888634, -0.008825386874377728, 0.013925113715231419, -0.004202894400805235, -0.033125173300504684, 0.004154479131102562, ...
wacv_2025_118a59d120
118a59d120
wacv
2,025
Uniform Attention Maps: Boosting Image Fidelity in Reconstruction and Editing
Text-guided image generation and editing using diffusion models have achieved remarkable advancements. Among these tuning-free methods have gained attention for their ability to perform edits without extensive model adjustments offering simplicity and efficiency. However existing tuning-free approaches often struggle w...
Wenyi Mo; Tianyu Zhang; Yalong Bai; Bing Su; Ji-Rong Wen
Gaoling School of Artificial Intelligence, Renmin University of China + Beijing Key Laboratory of Big Data Management and Analysis Methods; Du Xiaoman Technology; Du Xiaoman Technology; Gaoling School of Artificial Intelligence, Renmin University of China + Beijing Key Laboratory of Big Data Management and Analysis Metho...
Poster
main
https://github.com/Mowenyii/Uniform-Attention-Maps
https://openaccess.thecvf.com/content/WACV2025/html/Mo_Uniform_Attention_Maps_Boosting_Image_Fidelity_in_Reconstruction_and_Editing_WACV_2025_paper.html
0
2411.19652
Uniform Attention Maps: Boosting Image Fidelity in Reconstruction and Editing Text-guided image generation and editing using diffusion models have achieved remarkable advancements. Among these tuning-free methods have gained attention for their ability to perform edits without extensive model adjustments offering simpl...
[ -0.06159806251525879, -0.022716760635375977, -0.05076169967651367, 0.0055081844329833984, -0.011916399002075195, -0.026010870933532715, -0.00352361798286438, 0.02046668529510498, 0.005332678556442261, 0.02251875400543213, -0.044785499572753906, -0.02250075340270996, 0.004097387194633484, 0...
wacv_2025_a9f21f3674
a9f21f3674
wacv
2,025
Unifying Low-Resolution and High-Resolution Alignment by Event Cameras for Space-Time Video Super-Resolution
Event cameras deliver asynchronous pixel intensity changes which result in sparse event data that offers the advantages of high temporal resolution. These high temporal characteristics make researchers naturally incorporate event cameras into video frame interpolation (VFI) and video super-resolution (VSR). In this pap...
Hoonhee Cho; Jae-Young Kang; Taewoo Kim; Yuhwan Jeong; Kuk-Jin Yoon
KAIST; KAIST; KAIST; KAIST; KAIST
Poster
main
https://github.com/Chohoonhee/ESTNet
https://openaccess.thecvf.com/content/WACV2025/html/Cho_Unifying_Low-Resolution_and_High-Resolution_Alignment_by_Event_Cameras_for_Space-Time_WACV_2025_paper.html
0
Unifying Low-Resolution and High-Resolution Alignment by Event Cameras for Space-Time Video Super-Resolution Event cameras deliver asynchronous pixel intensity changes which result in sparse event data that offers the advantages of high temporal resolution. These high temporal characteristics make researchers naturally...
[ -0.03513257950544357, -0.013976696878671646, -0.046019501984119415, 0.03907658904790878, -0.02788069285452366, 0.07786238193511963, -0.013867645524442196, -0.008805864490568638, 0.004689191002398729, 0.03682287037372589, 0.00793345645070076, 0.016230415552854538, -0.01424932386726141, 0.05...
wacv_2025_29b0f053d5
29b0f053d5
wacv
2,025
Unleashing Potentials of Vision-Language Models for Zero-Shot HOI Detection
Human-Object Interaction (HOI) detection aims to localize human-object pairs and recognize their interactions as <human action object> triplets. Recent advancements in pre-trained vision-language model (VLM) have improved zero-shot HOI detection enabling identification of unseen triplets. However existing methods lever...
Moyuru Yamada; Nimish Dharamshi; Ayushi Kohli; Prasad Kasu; Ainulla Khan; Manu Ghulyani
Fujitsu Research of India Private Limited, Bangalore, KA, INDIA; ; ; ; ;
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Yamada_Unleashing_Potentials_of_Vision-Language_Models_for_Zero-Shot_HOI_Detection_WACV_2025_paper.html
0
Unleashing Potentials of Vision-Language Models for Zero-Shot HOI Detection Human-Object Interaction (HOI) detection aims to localize human-object pairs and recognize their interactions as <human action object> triplets. Recent advancements in pre-trained vision-language model (VLM) have improved zero-shot HOI detectio...
[ -0.03706030175089836, -0.04241841658949852, -0.03227892890572548, 0.060092758387327194, 0.013227848336100578, 0.0038348666857928038, 0.0020046422723680735, 0.026065001264214516, 0.001178483129478991, 0.01640922948718071, -0.03880913183093071, -0.02295803837478161, 0.021525487303733826, -0....
wacv_2025_08cad33315
08cad33315
wacv
2,025
Unsupervised Denoising for Signal-Dependent and Row-Correlated Imaging Noise
Accurate analysis of microscopy images is hindered by the presence of noise. This noise is usually signal-dependent and often additionally correlated along rows or columns of pixels. Current self- and unsupervised denoisers can address signal-dependent noise but none can reliably remove noise that is also row- or colum...
Benjamin Salmon; Alexander Krull
School of Computer, University of Birmingham; School of Computer, University of Birmingham
Poster
main
https://github.com/krulllab/COSDD
https://openaccess.thecvf.com/content/WACV2025/html/Salmon_Unsupervised_Denoising_for_Signal-Dependent_and_Row-Correlated_Imaging_Noise_WACV_2025_paper.html
2
2310.07887
Unsupervised Denoising for Signal-Dependent and Row-Correlated Imaging Noise Accurate analysis of microscopy images is hindered by the presence of noise. This noise is usually signal-dependent and often additionally correlated along rows or columns of pixels. Current self- and unsupervised denoisers can address signal-...
[ -0.01748400740325451, -0.017658663913607597, -0.033754613250494, 0.025150496512651443, 0.0177597813308239, 0.029360631480813026, 0.004444541409611702, 0.06423672288656235, -0.012115626595914364, 0.05335288122296333, -0.018210209906101227, -0.043645672500133514, 0.012317860499024391, 0.0136...
wacv_2025_739190fbdd
739190fbdd
wacv
2,025
Unsupervised Domain Adaptive Visual Question Answering in the Era of Multi-Modal Large Language Models
Unsupervised domain adaptation (UDA) for visual question answering (VQA) has attracted research interest. However with Multi-modal Large Language Models (MLLMs) showing great performance on VQA datasets UDA for VQA based on MLLMs remains unexplored. To fill this gap we propose the first systematic approach to Unsupervi...
Weixi Weng; Rui Zhang; Xiaojun Meng; Jieming Zhu; Qun Liu; Chun Yuan
Tsinghua Shenzhen International Graduate School, Tsinghua University; School of Computer Science & Tech, Huazhong University of Science and Technology; Huawei Noah Ark’s Lab; Huawei Noah Ark’s Lab; Huawei Noah Ark’s Lab; Tsinghua Shenzhen International Graduate School, Tsinghua University
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Weng_Unsupervised_Domain_Adaptive_Visual_Question_Answering_in_the_Era_of_WACV_2025_paper.html
0
Unsupervised Domain Adaptive Visual Question Answering in the Era of Multi-Modal Large Language Models Unsupervised domain adaptation (UDA) for visual question answering (VQA) has attracted research interest. However with Multi-modal Large Language Models (MLLMs) showing great performance on VQA datasets UDA for VQA ba...
[ -0.056683655828237534, 0.0009086570353247225, -0.03416826203465462, -0.02106338180601597, -0.04550866782665253, 0.026044338941574097, 0.02110014297068119, -0.007816058583557606, 0.015282899141311646, -0.005913737695664167, -0.034664519131183624, -0.008404215797781944, -0.0011160054709762335,...
wacv_2025_c47b8437e4
c47b8437e4
wacv
2,025
Unsupervised Single-Image Intrinsic Image Decomposition with LiDAR Intensity Enhanced Training
Unsupervised intrinsic image decomposition (IID) is the task of separating a natural image into albedo and shade without ground truth during training. Although a recent model employing light detection and ranging (LiDAR) intensity demonstrated impressive performance the necessity of LiDAR intensity during inference res...
Shogo Sato; Takuhiro Kaneko; Kazuhiko Murasaki; Taiga Yoshida; Ryuichi Tanida; Akisato Kimura
NTT Corporation; NTT Corporation; NTT Corporation; NTT Corporation; NTT Corporation; NTT Corporation
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Sato_Unsupervised_Single-Image_Intrinsic_Image_Decomposition_with_LiDAR_Intensity_Enhanced_Training_WACV_2025_paper.html
0
Unsupervised Single-Image Intrinsic Image Decomposition with LiDAR Intensity Enhanced Training Unsupervised intrinsic image decomposition (IID) is the task of separating a natural image into albedo and shade without ground truth during training. Although a recent model employing light detection and ranging (LiDAR) inte...
[ -0.04983542114496231, 0.0025537917390465736, -0.04655197635293007, 0.002971063135191798, -0.019864851608872414, -0.00024226818641182035, 0.040276944637298584, 0.012230838648974895, -0.0011246944777667522, 0.013772234320640564, -0.061436932533979416, -0.057387348264455795, 0.00038107362342998...
wacv_2025_0edbb136d7
0edbb136d7
wacv
2,025
Unsupervised Video Highlight Detection by Learning from Audio and Visual Recurrence
With the exponential growth of video content the need for automated video highlight detection to extract key moments or highlights from lengthy videos has become increasingly pressing. This technology has the potential to enhance user experiences by allowing quick access to relevant content across diverse domains. Exis...
Zahidul Islam; Sujoy Paul; Mrigank Rochan
;;
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Islam_Unsupervised_Video_Highlight_Detection_by_Learning_from_Audio_and_Visual_WACV_2025_paper.html
1
2407.13933
Unsupervised Video Highlight Detection by Learning from Audio and Visual Recurrence With the exponential growth of video content the need for automated video highlight detection to extract key moments or highlights from lengthy videos has become increasingly pressing. This technology has the potential to enhance user e...
[ -0.04786599427461624, -0.019735516980290413, -0.04186433181166649, 0.041901152580976486, 0.0018064809264615178, 0.025958096608519554, 0.013963982462882996, 0.0486023910343647, -0.03369029611349106, 0.021558105945587158, -0.003026143182069063, -0.05117979273200035, -0.016449332237243652, 0....
wacv_2025_46c9d34f9b
46c9d34f9b
wacv
2,025
User-in-the-Loop Evaluation of Multimodal LLMs for Activity Assistance
Our research investigates the capability of modern multimodal reasoning models powered by Large Language Models (LLMs) to facilitate vision-powered assistants for multi-step daily activities. Such assistants must be able to 1) encode relevant visual history from the assistant's sensors e.g. camera 2) forecast future ac...
Mrinal Verghese; Brian Chen; Hamid Eghbalzadeh; Tushar Nagarajan; Ruta P Desai
Carnegie Mellon University; Samsung Research America + Columbia University; Meta Reality Labs Research; Meta Fundamental AI Research; Meta Fundamental AI Research
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Verghese_User-in-the-Loop_Evaluation_of_Multimodal_LLMs_for_Activity_Assistance_WACV_2025_paper.html
1
2408.03160
User-in-the-Loop Evaluation of Multimodal LLMs for Activity Assistance Our research investigates the capability of modern multimodal reasoning models powered by Large Language Models (LLMs) to facilitate vision-powered assistants for multi-step daily activities. Such assistants must be able to 1) encode relevant visual...
[ -0.03548299893736839, -0.008824667893350124, -0.0077647864818573, 0.009594234637916088, -0.01882903091609478, 0.0026520085521042347, 0.015400543808937073, 0.002414687303826213, 0.004375468473881483, -0.023907247930765152, -0.01794425956904888, -0.007930681109428406, -0.00030529205105267465, ...
wacv_2025_d05ed56c5d
d05ed56c5d
wacv
2,025
Utilizing Uncertainty in 2D Pose Detectors for Probabilistic 3D Human Mesh Recovery
Monocular 3D human pose and shape estimation is an inherently ill-posed problem due to depth ambiguities occlusions and truncations. Recent probabilistic approaches learn a distribution over plausible 3D human meshes by maximizing the likelihood of the ground-truth pose given an image. We show that this objective funct...
Tom Wehrbein; Marco Rudolph; Bodo Rosenhahn; Bastian Wandt
Leibniz University Hannover; Leibniz University Hannover; Leibniz University Hannover; Linköping University
Poster
main
https://github.com/twehrbein/humr
https://openaccess.thecvf.com/content/WACV2025/html/Wehrbein_Utilizing_Uncertainty_in_2D_Pose_Detectors_for_Probabilistic_3D_Human_WACV_2025_paper.html
2
2411.16289
Utilizing Uncertainty in 2D Pose Detectors for Probabilistic 3D Human Mesh Recovery Monocular 3D human pose and shape estimation is an inherently ill-posed problem due to depth ambiguities occlusions and truncations. Recent probabilistic approaches learn a distribution over plausible 3D human meshes by maximizing the l...
[ -0.02680206298828125, -0.03284511715173721, -0.046068958938121796, 0.009922545403242111, -0.02143045887351036, 0.01034220214933157, 0.03170737996697426, -0.008640261366963387, -0.0077962856739759445, 0.009288397617638111, -0.056812167167663574, 0.009036603383719921, 0.024936921894550323, 0...
wacv_2025_f761237094
f761237094
wacv
2,025
V-MIND: Building Versatile Monocular Indoor 3D Detector with Diverse 2D Annotations
The field of indoor monocular 3D object detection is gaining significant attention fueled by the increasing demand in VR/AR and robotic applications. However its advancement is impeded by the limited availability and diversity of 3D training data owing to the labor-intensive nature of 3D data collection and annotation ...
Jin-Cheng Jhang; Tao Tu; Fu-En Wang; Ke Zhang; Min Sun; Cheng-Hao Kuo
National Tsing Hua University; Cornell University+Amazon; Amazon; Amazon; National Tsing Hua University+Amazon; Amazon
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Jhang_V-MIND_Building_Versatile_Monocular_Indoor_3D_Detector_with_Diverse_2D_WACV_2025_paper.html
2
V-MIND: Building Versatile Monocular Indoor 3D Detector with Diverse 2D Annotations The field of indoor monocular 3D object detection is gaining significant attention fueled by the increasing demand in VR/AR and robotic applications. However its advancement is impeded by the limited availability and diversity of 3D tra...
[ -0.03604044020175934, -0.037620048969984055, -0.068703792989254, 0.01683097705245018, 0.017856813967227936, 0.03275413438677788, -0.020589349791407585, 0.004627611022442579, 0.008964719250798225, 0.002612251089885831, -0.05523175001144409, -0.022041862830519676, -0.021097728982567787, -0.0...
wacv_2025_739559590d
739559590d
wacv
2,025
VADet: Multi-Frame LiDAR 3D Object Detection using Variable Aggregation
Input aggregation is a simple technique used by state-of-the-art LiDAR 3D object detectors to improve detection. However increasing aggregation is known to have diminishing returns and even performance degradation due to objects responding differently to the number of aggregated frames. To address this limitation we pr...
Chengjie Huang; Vahdat Abdelzad; Sean Sedwards; Krzysztof Czarnecki
University of Waterloo; University of Waterloo; University of Waterloo; University of Waterloo
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Huang_VADet_Multi-Frame_LiDAR_3D_Object_Detection_using_Variable_Aggregation_WACV_2025_paper.html
0
2411.13186
VADet: Multi-Frame LiDAR 3D Object Detection using Variable Aggregation Input aggregation is a simple technique used by state-of-the-art LiDAR 3D object detectors to improve detection. However increasing aggregation is known to have diminishing returns and even performance degradation due to objects responding differen...
[ 0.00237132771871984, -0.018879417330026627, -0.0470617339015007, 0.018222741782665253, -0.00609704852104187, -0.0059648011811077595, 0.003577512688934803, 0.011929602362215519, 0.029751041904091835, 0.0032309340313076973, -0.03721160441637039, -0.0337093360722065, 0.013689856976270676, 0.0...
wacv_2025_e6faa9802c
e6faa9802c
wacv
2,025
VG-SSL: Benchmarking Self-Supervised Representation Learning Approaches for Visual Geo-Localization
Visual Geo-localization (VG) is a critical research area for identifying geo-locations from visual inputs particularly in autonomous navigation for robotics and vehicles. Current VG methods often learn feature extractors from geo-labeled images to create dense geographically relevant representations. Recent advances in...
Jiuhong Xiao; Gao Zhu; Giuseppe Loianno
New York University; New York University; New York University
Poster
main
https://github.com/arplaboratory/VG-SSL
https://openaccess.thecvf.com/content/WACV2025/html/Xiao_VG-SSL_Benchmarking_Self-Supervised_Representation_Learning_Approaches_for_Visual_Geo-Localization_WACV_2025_paper.html
0
VG-SSL: Benchmarking Self-Supervised Representation Learning Approaches for Visual Geo-Localization Visual Geo-localization (VG) is a critical research area for identifying geo-locations from visual inputs particularly in autonomous navigation for robotics and vehicles. Current VG methods often learn feature extractors...
[ -0.09150578826665878, -0.02543642930686474, 0.009055259637534618, 0.002811896614730358, -0.004146356135606766, 0.035149842500686646, 0.012255239300429821, -0.004516282584518194, 0.010357946157455444, 0.025273026898503304, 0.019426822662353516, -0.01855533942580223, -0.004582097753882408, 0...
wacv_2025_cbad6fa43e
cbad6fa43e
wacv
2,025
VHS: High-Resolution Iterative Stereo Matching with Visual Hull Priors
We present a stereo-matching method for depth estimation from high-resolution images using visual hulls as priors and a memory-efficient technique for the correlation computation. Our method uses object masks extracted from supplementary views of the scene to guide the disparity estimation effectively reducing the sear...
Markus Plack; Hannah Dröge; Leif Van Holland; Matthias B. Hullin
University of Bonn; University of Bonn; University of Bonn; University of Bonn
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Plack_VHS_High-Resolution_Iterative_Stereo_Matching_with_Visual_Hull_Priors_WACV_2025_paper.html
0
VHS: High-Resolution Iterative Stereo Matching with Visual Hull Priors We present a stereo-matching method for depth estimation from high-resolution images using visual hulls as priors and a memory-efficient technique for the correlation computation. Our method uses object masks extracted from supplementary views of th...
[ -0.05736342445015907, -0.035323210060596466, -0.026713179424405098, 0.04010656476020813, -0.01237232331186533, 0.024891825392842293, 0.060748565942049026, 0.004893736448138952, 0.004935130942612886, 0.024542273953557014, -0.028037799522280693, -0.030521463602781296, 0.0021007652394473553, ...
wacv_2025_2edf69c0cb
2edf69c0cb
wacv
2,025
VIIS: Visible and Infrared Information Synthesis for Severe Low-Light Image Enhancement
Images captured in severe low-light circumstances often suffer from significant information absence. Existing singular modality image enhancement methods struggle to restore image regions lacking valid information. By leveraging light-impervious infrared images visible and infrared image fusion methods have the potenti...
Chen Zhao; Mengyuan Yu; Fan Yang; Peiguang Jing
Tianjin University, Tianjin, China; Southeast University, Nanjing, China; Tianjin University, Tianjin, China; Tianjin University, Tianjin, China
Poster
main
https://github.com/Chenz418/VIIS
https://openaccess.thecvf.com/content/WACV2025/html/Zhao_VIIS_Visible_and_Infrared_Information_Synthesis_for_Severe_Low-Light_Image_WACV_2025_paper.html
0
2412.13655
VIIS: Visible and Infrared Information Synthesis for Severe Low-Light Image Enhancement Images captured in severe low-light circumstances often suffer from significant information absence. Existing singular modality image enhancement methods struggle to restore image regions lacking valid information. By leveraging lig...
[ -0.048336051404476166, 0.004008503630757332, -0.01982489414513111, 0.0650346428155899, -0.03964978829026222, -0.0365234836935997, 0.02955949865281582, 0.04792420193552971, 0.035175614058971405, 0.044479649513959885, -0.042832255363464355, -0.03156258165836334, -0.015313279815018177, 0.0367...
wacv_2025_2efdbc1958
2efdbc1958
wacv
2,025
VILLS : Video-Image Learning to Learn Semantics for Person Re-Identification
Person Re-identification is a research area with significant real world applications. Despite recent progress existing methods face challenges in robust re-identification in the wild e.g. by focusing only on a particular modality and on unreliable patterns such as clothing. A generalized method is highly desired but re...
Siyuan Huang; Ram Prabhakar Kathirvel; Yuxiang Guo; Rama Chellappa; Cheng Peng
Johns Hopkins University; Johns Hopkins University; Johns Hopkins University; Johns Hopkins University; Johns Hopkins University
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Huang_VILLS__Video-Image_Learning_to_Learn_Semantics_for_Person_Re-Identification_WACV_2025_paper.html
1
VILLS : Video-Image Learning to Learn Semantics for Person Re-Identification Person Re-identification is a research area with significant real world applications. Despite recent progress existing methods face challenges in robust re-identification in the wild e.g. by focusing only on a particular modality and on unreli...
[ -0.08191288262605667, 0.021946260705590248, -0.024900922551751137, 0.018229704350233078, -0.01566528156399727, -0.013491096906363964, 0.01845269836485386, 0.018963724374771118, 0.010489977896213531, 0.018499154597520828, -0.011242580600082874, -0.026257464662194252, -0.016724498942494392, ...
wacv_2025_38f9bd97f9
38f9bd97f9
wacv
2,025
VISIONARY: Novel Spatial-Spectral Attention Mechanism for Hyperspectral Image Denoising
Image denoising mitigates noise from the captured images and thereby enhances the efficacy of high-demand vision applications such as classification and segmentation. Hyperspectral Images (HSIs) with their multiple spectral bands provide valuable information and make them highly applicable in real-world applications. C...
Aditya Dixit; Nischit Hosamani; Puneet Gupta; Ankur Garg
IIT Indore, India; IIT Indore, India; IIT Indore, India; SAC, Ahmedabad, India
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Dixit_VISIONARY_Novel_Spatial-Spectral_Attention_Mechanism_for_Hyperspectral_Image_Denoising_WACV_2025_paper.html
0
VISIONARY: Novel Spatial-Spectral Attention Mechanism for Hyperspectral Image Denoising Image denoising mitigates noise from the captured images and thereby enhances the efficacy of high-demand vision applications such as classification and segmentation. Hyperspectral Images (HSIs) with their multiple spectral bands pr...
[ -0.051762841641902924, -0.023947222158312798, -0.022215653210878372, 0.015086750499904156, -0.016864370554685593, 0.005286809988319874, 0.009855203330516815, 0.0005540673155337572, 0.05529966205358505, 0.029234033077955246, -0.011503877118229866, 0.004347342066466808, 0.020226191729307175, ...
wacv_2025_34fc534cf2
34fc534cf2
wacv
2,025
VLTP: Vision-Language Guided Token Pruning for Task-Oriented Segmentation
Vision Transformers (ViTs) have emerged as the backbone of many segmentation models consistently achieving state-of-the-art (SOTA) performance. However their success comes at a significant computational cost. Image token pruning is one of the most effective strategies to address this complexity. However previous approa...
Hanning Chen; Yang Ni; Wenjun Huang; Yezi Liu; SungHeon Jeong; Fei Wen; Nathaniel Bastian; Hugo Latapie; Mohsen Imani
University of California, Irvine, CA, USA; University of California, Irvine, CA, USA; University of California, Irvine, CA, USA; University of California, Irvine, CA, USA; University of California, Irvine, CA, USA; Texas A&M University, College Station, TX, USA; United States Military Academy, West Point, NY, USA; Cisc...
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Chen_VLTP_Vision-Language_Guided_Token_Pruning_for_Task-Oriented_Segmentation_WACV_2025_paper.html
5
2409.08464
VLTP: Vision-Language Guided Token Pruning for Task-Oriented Segmentation Vision Transformers (ViTs) have emerged as the backbone of many segmentation models consistently achieving state-of-the-art (SOTA) performance. However their success comes at a significant computational cost. Image token pruning is one of the mos...
[ -0.04077553376555443, -0.014111104421317577, -0.0029083085246384144, 0.03224874660372734, -0.026272708550095558, 0.0019961881916970015, 0.024432525038719177, -0.004577682353556156, 0.027274787425994873, 0.011414598673582077, -0.01380137074738741, -0.02366730011999607, 0.0038443421944975853, ...
wacv_2025_36ee09cd6c
36ee09cd6c
wacv
2,025
VM-Gait: Multi-Modal 3D Representation Based on Virtual Marker for Gait Recognition
Gait recognition plays a vital role in biometric applications by analyzing the unique characteristics of an individual's walking pattern. Methods based on 2D representations such as silhouettes and skeletons are increasingly being developed to learn the shape features and joint dynamic movements. Nevertheless the effec...
Zhao-Yang Wang; Jiang Liu; Jieneng Chen; Rama Chellappa
;;;
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Wang_VM-Gait_Multi-Modal_3D_Representation_Based_on_Virtual_Marker_for_Gait_WACV_2025_paper.html
0
VM-Gait: Multi-Modal 3D Representation Based on Virtual Marker for Gait Recognition Gait recognition plays a vital role in biometric applications by analyzing the unique characteristics of an individual's walking pattern. Methods based on 2D representations such as silhouettes and skeletons are increasingly being devel...
[ -0.02992388606071472, -0.03992334380745888, -0.03336876258254051, 0.03042665123939514, -0.01560436561703682, -0.002214730717241764, 0.0006674450705759227, 0.0029188357293605804, 0.03621777147054672, -0.004399202298372984, -0.014812975190579891, 0.0013139416696503758, 0.004308424890041351, ...
wacv_2025_24be58e721
24be58e721
wacv
2,025
VMAs: Video-to-Music Generation via Semantic Alignment in Web Music Videos
We present a framework for learning to generate background music from video inputs. Unlike existing works that rely on symbolic musical annotations which are limited in quantity and diversity our method leverages large-scale web videos accompanied by background music. This enables our model to learn to generate realist...
Yan-Bo Lin; Yu Tian; Linjie Yang; Gedas Bertasius; Heng Wang
UNC Chapel Hill; ByteDance Inc.; ByteDance Inc.; UNC Chapel Hill; ByteDance Inc.
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Lin_VMAs_Video-to-Music_Generation_via_Semantic_Alignment_in_Web_Music_Videos_WACV_2025_paper.html
7
2409.07450
VMAs: Video-to-Music Generation via Semantic Alignment in Web Music Videos We present a framework for learning to generate background music from video inputs. Unlike existing works that rely on symbolic musical annotations which are limited in quantity and diversity our method leverages large-scale web videos accompani...
[ -0.027717214077711105, 0.0014669434167444706, -0.05077273026108742, 0.03537829592823982, -0.025476709008216858, -0.013804401271045208, -0.018330581486225128, 0.029795100912451744, -0.01083211787045002, 0.03387860208749771, -0.00900719128549099, -0.007728838361799717, -0.006296902894973755, ...
wacv_2025_7aa6d1d0ca
7aa6d1d0ca
wacv
2,025
VaLID: Variable-Length Input Diffusion for Novel View Synthesis
Novel View Synthesis (NVS) which tries to produce a realistic image at the target view given source view images and their corresponding poses is a fundamental problem in 3D Vision. As this task is heavily under-constrained some recent work like Zero123 [18] tries to solve this problem with generative modeling specifica...
Shijie Li; Farhad G. Zanjani; Haitam Ben Yahia; Yuki Asano; Juergen Gall; Amirhossein Habibian
University of Bonn+Qualcomm AI Research; Qualcomm AI Research; Qualcomm AI Research; Qualcomm AI Research; University of Bonn; Qualcomm AI Research
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Li_VaLID_Variable-Length_Input_Diffusion_for_Novel_View_Synthesis_WACV_2025_paper.html
5
2312.08892
VaLID: Variable-Length Input Diffusion for Novel View Synthesis Novel View Synthesis (NVS) which tries to produce a realistic image at the target view given source view images and their corresponding poses is a fundamental problem in 3D Vision. As this task is heavily under-constrained some recent work like Zero123 [18...
[ -0.017870232462882996, -0.023245980963110924, -0.053574007004499435, 0.04282251000404358, -0.007325144484639168, 0.013650363311171532, 0.0034584589302539825, 0.033061765134334564, 0.023869788274168968, 0.027080558240413666, -0.021209435537457466, -0.0404006727039814, 0.008930529467761517, ...
wacv_2025_ea04adb8c0
ea04adb8c0
wacv
2,025
VerA: Versatile Anonymization Applicable to Clinical Facial Photographs
The demand for privacy in facial image dissemination is gaining ground internationally echoed by the proliferation of regulations such as GDPR DPDPA CCPA PIPL and APPI. While recent advances in anonymization surpass pixelation or blur methods additional constraints to the task pose challenges. Largely unaddressed by cu...
Majed El Helou; Doruk Cetin; Petar Stamenkovic; Niko Benjamin Huber; Fabio Zünd
ETH Zurich, Switzerland; Align Technology, Switzerland; ETH Zurich, Switzerland; Align Technology, Switzerland; ETH Zurich, Switzerland
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Helou_VerA_Versatile_Anonymization_Applicable_to_Clinical_Facial_Photographs_WACV_2025_paper.html
1
2312.02124
VerA: Versatile Anonymization Applicable to Clinical Facial Photographs The demand for privacy in facial image dissemination is gaining ground internationally echoed by the proliferation of regulations such as GDPR DPDPA CCPA PIPL and APPI. While recent advances in anonymization surpass pixelation or blur methods addit...
[ -0.07084488123655319, 0.008215500973165035, 0.006810944993048906, -0.0021160440519452095, -0.05666116625070572, 0.022638678550720215, 0.02460966259241104, 0.016550734639167786, 0.045608922839164734, 0.008620750159025192, -0.0018823351711034775, 0.00013268447946757078, 0.03149889409542084, ...
wacv_2025_f9a72eb06c
f9a72eb06c
wacv
2,025
VideoGameBunny: Towards Vision Assistants for Video Games
Large multimodal models known as LMMs hold substantial promise across various domains from personal assistance in daily tasks to sophisticated applications like medical diagnostics. However their capabilities have limitations in the video game domain including challenges with scene understanding hallucinations and inac...
Mohammad Reza Taesiri; Cor-Paul Bezemer
University of Alberta; University of Alberta
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Taesiri_VideoGameBunny_Towards_Vision_Assistants_for_Video_Games_WACV_2025_paper.html
3
2407.15295
VideoGameBunny: Towards Vision Assistants for Video Games Large multimodal models known as LMMs hold substantial promise across various domains from personal assistance in daily tasks to sophisticated applications like medical diagnostics. However their capabilities have limitations in the video game domain including c...
[ -0.06669031828641891, -0.0128571055829525, -0.031167592853307724, 0.007621480152010918, -0.03823047876358032, 0.02198394574224949, -0.010083075612783432, -0.019863186404109, -0.010935166850686073, -0.007664084434509277, -0.024805312976241112, 0.027834970504045486, -0.016265468671917915, 0....
wacv_2025_1d4f6b0c49
1d4f6b0c49
wacv
2,025
VioPose: Violin Performance 4D Pose Estimation by Hierarchical Audiovisual Inference
Musicians delicately control their bodies to generate music. Sometimes their motions are too subtle to be captured by the human eye. To analyze how they move to produce the music we need to estimate precise 4D human pose (3D pose over time). However current state-of-the-art (SoTA) visual pose estimation algorithms stru...
Seong Jong Yoo; Snehesh Shrestha; Irina Muresanu; Cornelia Fermuller
University of Maryland, College Park; University of Maryland, College Park; University of Maryland, College Park; University of Maryland, College Park
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Yoo_VioPose_Violin_Performance_4D_Pose_Estimation_by_Hierarchical_Audiovisual_Inference_WACV_2025_paper.html
0
2411.13607
VioPose: Violin Performance 4D Pose Estimation by Hierarchical Audiovisual Inference Musicians delicately control their bodies to generate music. Sometimes their motions are too subtle to be captured by the human eye. To analyze how they move to produce the music we need to estimate precise 4D human pose (3D pose over ...
[ -0.025903480127453804, 0.010447988286614418, -0.045029886066913605, 0.04326031729578972, 0.011822227388620377, 0.027560098096728325, 0.03358416259288788, 0.03702917322516441, 0.01581317000091076, 0.038817569613456726, -0.0008336035534739494, -0.01489073596894741, 0.01811925880610943, 0.000...
wacv_2025_165db0d03a
165db0d03a
wacv
2,025
VipDiff: Towards Coherent and Diverse Video Inpainting via Training-Free Denoising Diffusion Models
Recent video inpainting methods have achieved encouraging improvements by leveraging optical flow to guide pixel propagation from reference frames either in the image space or feature space. However they would produce severe artifacts when the masked area is too large and no pixel correspondences could be found. Recent...
Chaohao Xie; Kai Han; Kwan-Yee K. Wong
The University of Hong Kong; The University of Hong Kong; The University of Hong Kong
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Xie_VipDiff_Towards_Coherent_and_Diverse_Video_Inpainting_via_Training-Free_Denoising_WACV_2025_paper.html
0
2501.12267
VipDiff: Towards Coherent and Diverse Video Inpainting via Training-Free Denoising Diffusion Models Recent video inpainting methods have achieved encouraging improvements by leveraging optical flow to guide pixel propagation from reference frames either in the image space or feature space. However they would produce se...
[ -0.09267885982990265, -0.03447961434721947, -0.023023070767521858, 0.029915329068899155, -0.0032101236283779144, 0.025076083838939667, 0.019576942548155785, 0.026414208114147186, 0.012638859450817108, 0.034864556044340134, -0.02117169462144375, -0.014884342439472675, 0.0071534663438797, 0....
wacv_2025_7019e3cd9e
7019e3cd9e
wacv
2,025
Vision-Aware Text Features in Referring Image Segmentation: From Object Understanding to Context Understanding
Referring image segmentation is a challenging task that involves generating pixel-wise segmentation masks based on natural language descriptions. The complexity of this task increases with the intricacy of the sentences provided. Existing methods have relied mostly on visual features to generate the segmentation masks ...
Hai Nguyen-Truong; E-Ro Nguyen; Tuan-Anh Vu; Minh-Triet Tran; Binh-Son Hua; Sai-Kit Yeung
The Hong Kong University of Science and Technology; Stony Brook University + University of Science, VNU-HCM, Ho Chi Minh City + Vietnam National University, Ho Chi Minh City; The Hong Kong University of Science and Technology; University of Science, VNU-HCM, Ho Chi Minh City + Vietnam National University, Ho Chi Minh C...
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Nguyen-Truong_Vision-Aware_Text_Features_in_Referring_Image_Segmentation_From_Object_Understanding_WACV_2025_paper.html
1
2404.08590
Vision-Aware Text Features in Referring Image Segmentation: From Object Understanding to Context Understanding Referring image segmentation is a challenging task that involves generating pixel-wise segmentation masks based on natural language descriptions. The complexity of this task increases with the intricacy of the...
[ -0.009235250763595104, -0.032858818769454956, 0.004432279616594315, 0.03576943278312683, -0.010782086290419102, -0.033316463232040405, 0.03271237388253212, 0.011221423745155334, 0.00822843424975872, 0.009294743649661541, -0.00913914479315281, -0.05191510543227196, -0.009958327747881413, 0....
wacv_2025_65df781842
65df781842
wacv
2,025
Vision-Based Landing Guidance through Tracking and Orientation Estimation
Fixed-wing aerial vehicles are equipped with functionalities such as ILS (instrument landing system) PAR (precision approach radar) and DGPS (differential global positioning system) enabling fully automated landings. However these systems impose significant costs on airport operations due to high installation and maint...
João P. K. Ferreira; João P. Pinto; Júlia Moura; Yi Li; Cristiano L. Castro; Plamen Angelov
;;;;;
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Ferreira_Vision-Based_Landing_Guidance_through_Tracking_and_Orientation_Estimation_WACV_2025_paper.html
0
Vision-Based Landing Guidance through Tracking and Orientation Estimation Fixed-wing aerial vehicles are equipped with functionalities such as ILS (instrument landing system) PAR (precision approach radar) and DGPS (differential global positioning system) enabling fully automated landings. However these systems impose ...
[ -0.010748263448476791, 0.00425376184284687, -0.026050874963402748, 0.0335746593773365, -0.014992916025221348, 0.03534174710512161, 0.0301497895270586, 0.0005638853181153536, 0.012032589875161648, 0.029402876272797585, -0.02779974602162838, -0.027071049436926842, 0.015584982000291348, 0.045...
wacv_2025_0698b178b7
0698b178b7
wacv
2,025
Visual Robustness Benchmark for Visual Question Answering (VQA)
Can Visual Question Answering (VQA) systems maintain their performance when deployed in the real world? Or are they susceptible to realistic corruption effects e.g. image blur which can be detrimental in sensitive applications such as medical VQA? While linguistic robustness has been thoroughly explored within the VQA ...
Farhan Ishmam; Ishmam Tashdeed; Talukder Asir Saadat; Hamjajul Ashmafee; Abu Raihan Mostofa Kamal; Azam Hossain
Department of Computer Science and Engineering, Islamic University of Technology; Department of Computer Science and Engineering, Islamic University of Technology; Department of Computer Science and Engineering, Islamic University of Technology; Department of Computer Science and Engineering, Islamic University of Tech...
Poster
main
https://github.com/ishmamt/Visual-Robustness
https://openaccess.thecvf.com/content/WACV2025/html/Ishmam_Visual_Robustness_Benchmark_for_Visual_Question_Answering_VQA_WACV_2025_paper.html
2
2407.03386
Visual Robustness Benchmark for Visual Question Answering (VQA) Can Visual Question Answering (VQA) systems maintain their performance when deployed in the real world? Or are they susceptible to realistic corruption effects e.g. image blur which can be detrimental in sensitive applications such as medical VQA? While li...
[ -0.053409792482852936, 0.009558181278407574, -0.006044625770300627, -0.005211852490901947, -0.026592610403895378, -0.007092610467225313, 0.013998080976307392, 0.031944818794727325, 0.007013075985014439, -0.000047808451199671254, -0.022868523374199867, -0.0014058902161195874, 0.00921197235584...
wacv_2025_2c1908ba30
2c1908ba30
wacv
2,025
VisualFusion: Enhancing Blog Content with Advanced Infographic Pipeline
Infographics represent a key component of any blog or article facilitating effective communication of ideas while fostering reader engagement. However many content creators possess limited expertise in crafting visually striking infographics. This gap is effectively addressed by our proposed pipeline designed to aid wr...
Anurag Deo; Savita Bhat; Shirish Karande
Indian Institute of Technology Patna + TCS Research; TCS Research; TCS Research
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Deo_VisualFusion_Enhancing_Blog_Content_with_Advanced_Infographic_Pipeline_WACV_2025_paper.html
0
VisualFusion: Enhancing Blog Content with Advanced Infographic Pipeline Infographics represent a key component of any blog or article facilitating effective communication of ideas while fostering reader engagement. However many content creators possess limited expertise in crafting visually striking infographics. This ...
[ -0.07581940293312073, -0.020844830200076103, 0.006931456737220287, 0.02989104762673378, -0.0195420291274786, -0.05739668384194374, -0.024477997794747353, 0.015092317946255207, -0.011670169420540333, 0.029872698709368706, -0.008775563910603523, -0.02244122140109539, -0.018037384375929832, 0...
wacv_2025_5788294c15
5788294c15
wacv
2,025
Volumetric Conditioning Module to Control Pretrained Diffusion Models for 3D Medical Images
Spatial control methods using additional modules on pretrained diffusion models have gained attention for enabling conditional generation in natural images. These methods guide the generation process with new conditions while leveraging the capabilities of large models. They could be beneficial as training strategies i...
Suhyun Ahn; Wonjung Park; Jihoon Cho; Jinah Park
KAIST; KAIST; KAIST; KAIST
Poster
main
https://github.com/SSTDV-Project/VCM.git
https://openaccess.thecvf.com/content/WACV2025/html/Ahn_Volumetric_Conditioning_Module_to_Control_Pretrained_Diffusion_Models_for_3D_WACV_2025_paper.html
0
2410.21826
Volumetric Conditioning Module to Control Pretrained Diffusion Models for 3D Medical Images Spatial control methods using additional modules on pretrained diffusion models have gained attention for enabling conditional generation in natural images. These methods guide the generation process with new conditions while le...
[ -0.05607078596949577, -0.01013910211622715, -0.03759954869747162, -0.024640481919050217, -0.009500274434685707, -0.022705748677253723, -0.007684180047363043, 0.02465873397886753, 0.032945238053798676, 0.03336503729224205, -0.03880419582128525, 0.009285811334848404, 0.006954091601073742, 0....
wacv_2025_9843cd7088
9843cd7088
wacv
2,025
VortSDF: 3D Modeling with Centroidal Voronoi Tesselation on Signed Distance Field
Volumetric shape representations have become ubiquitous in multi-view reconstruction tasks. They often build on regular voxel grids as discrete representations of 3D shape functions such as SDF or radiance fields either as the full shape model or as sampled instantiations of continuous representations as with neural ne...
Diego Thomas; Briac Toussaint; Jean-Sebastien Franco; Edmond Boyer
Kyushu University (Japan); INRIA Grenoble Rhone-Alpes-LJK (France); INRIA Grenoble Rhone-Alpes-LJK (France); Meta Reality Labs
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Thomas_VortSDF_3D_Modeling_with_Centroidal_Voronoi_Tesselation_on_Signed_Distance_WACV_2025_paper.html
0
2407.19837
VortSDF: 3D Modeling with Centroidal Voronoi Tesselation on Signed Distance Field Volumetric shape representations have become ubiquitous in multi-view reconstruction tasks. They often build on regular voxel grids as discrete representations of 3D shape functions such as SDF or radiance fields either as the full shape ...
[ -0.048975955694913864, -0.011271807365119457, -0.013766462914645672, -0.009281585924327374, -0.02454300783574581, 0.017673533409833908, 0.0038726774509996176, 0.016362003982067108, 0.059908416122198105, 0.011381865479052067, -0.003923120908439159, -0.009960278868675232, -0.01793033629655838,...
wacv_2025_91a31dabb2
91a31dabb2
wacv
2,025
WAFFLE: Multimodal Floorplan Understanding in the Wild
Buildings are a central feature of human culture and are increasingly being analyzed with computational methods. However recent works on computational building understanding have largely focused on natural imagery of buildings neglecting the fundamental element defining a building's structure - its floorplan. Conversel...
Keren Ganon; Morris Alper; Rachel Mikulinsky; Hadar Averbuch-Elor
;;;
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Ganon_WAFFLE_Multimodal_Floorplan_Understanding_in_the_Wild_WACV_2025_paper.html
0
2412.00955
WAFFLE: Multimodal Floorplan Understanding in the Wild Buildings are a central feature of human culture and are increasingly being analyzed with computational methods. However recent works on computational building understanding have largely focused on natural imagery of buildings neglecting the fundamental element def...
[ -0.0548609159886837, 0.022346528246998787, -0.006677400320768356, 0.001107960008084774, -0.007578470278531313, -0.02293459326028824, 0.02181537076830864, 0.017983453348279, -0.003189312294125557, 0.012643430382013321, -0.004797010216861963, -0.009807432070374489, -0.028682470321655273, 0.0...
wacv_2025_b9097c593a
b9097c593a
wacv
2,025
WARLearn: Weather-Adaptive Representation Learning
This paper introduces WARLearn a novel framework designed for adaptive representation learning in challenging and adversarial weather conditions. Leveraging the in-variance principal used in Barlow Twins we demonstrate the capability to port the existing models initially trained on clear weather data to effectively han...
Shubham Agarwal; Raz Birman; Ofer Hadar
Ben-Gurion University of the Negev, Israel; Ben-Gurion University of the Negev, Israel; Ben-Gurion University of the Negev, Israel
Poster
main
https://github.com/ShubhamAgarwal12/WARLearn
https://openaccess.thecvf.com/content/WACV2025/html/Agarwal_WARLearn_Weather-Adaptive_Representation_Learning_WACV_2025_paper.html
0
2411.14095
WARLearn: Weather-Adaptive Representation Learning This paper introduces WARLearn a novel framework designed for adaptive representation learning in challenging and adversarial weather conditions. Leveraging the in-variance principal used in Barlow Twins we demonstrate the capability to port the existing models initial...
[ -0.023182645440101624, -0.004535537213087082, -0.037012167274951935, 0.0008671883260831237, -0.04170692712068558, -0.046438079327344894, 0.008575213141739368, 0.03402790054678917, 0.013438291847705841, -0.00820218026638031, -0.03559282049536705, -0.01815124787390232, -0.0008125980966724455, ...
wacv_2025_ce1601e2e5
ce1601e2e5
wacv
2,025
WINE : Wavelet-Guided GAN Inversion and Editing for High-Fidelity Refinement
Recent advanced GAN inversion models aim to convey high-fidelity information from original images to generators through methods using generator tuning or high-dimensional feature learning. Despite these efforts accurately reconstructing image-specific details remains as a challenge due to the inherent limitations both ...
Chaewon Kim; Seung Jun Moon; Gyeong-Moon Park
KRAFTON; Klleon AI Research; Kyung Hee University
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Kim_WINE__Wavelet-Guided_GAN_Inversion_and_Editing_for_High-Fidelity_Refinement_WACV_2025_paper.html
0
2210.09655
WINE : Wavelet-Guided GAN Inversion and Editing for High-Fidelity Refinement Recent advanced GAN inversion models aim to convey high-fidelity information from original images to generators through methods using generator tuning or high-dimensional feature learning. Despite these efforts accurately reconstructing image-...
[ -0.06843948364257812, -0.015594159252941608, -0.05054850876331329, 0.0024967393837869167, -0.01746322773396969, -0.030946549028158188, 0.02572060003876686, 0.03431273251771927, 0.000223317532800138, 0.02687365561723709, -0.02235441841185093, -0.010312416590750217, -0.011539864353835583, 0....
wacv_2025_1320990900
1320990900
wacv
2,025
Wavelength- and Depth-Aware Deep Image Prior for Blind Hyperspectral Imagery Deblurring with Coarse Depth Guidance
Hyperspectral imagery (HSI) provides detailed spectral information enabling precise analysis of materials. However HSI imaging suffers from blurring degradation which results in the loss of fine details and hinders subsequent applications. The degree of blurriness is highly related to wavelength and depth existing debl...
Jiahuan Li; Xiaoyu Dong; Wei He; Naoto Yokoya
;;;
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Li_Wavelength-_and_Depth-Aware_Deep_Image_Prior_for_Blind_Hyperspectral_Imagery_WACV_2025_paper.html
0
Wavelength- and Depth-Aware Deep Image Prior for Blind Hyperspectral Imagery Deblurring with Coarse Depth Guidance Hyperspectral imagery (HSI) provides detailed spectral information enabling precise analysis of materials. However HSI imaging suffers from blurring degradation which results in the loss of fine details an...
[ -0.03799737989902496, -0.025727391242980957, -0.01822592504322529, 0.0020909395534545183, -0.01709505170583725, -0.0042478470131754875, 0.016831180080771446, -0.03479323536157608, 0.02094002440571785, 0.043840229511260986, -0.04580041021108627, -0.02346564270555973, 0.036150284111499786, 0...
wacv_2025_bbc20c33c8
bbc20c33c8
wacv
2,025
WeedsGalore: A Multispectral and Multitemporal UAV-Based Dataset for Crop and Weed Segmentation in Agricultural Maize Fields
Weeds are one of the major reasons for crop yield loss but current weeding practices fail to manage weeds in an efficient and targeted manner. Effective weed management is especially important for crops with high worldwide production such as maize to maximize crop yield for meeting increasing global demands. Advances i...
Ekin Celikkan; Timo Kunzmann; Yertay Yeskaliyev; Sibylle Itzerott; Nadja Klein; Martin Herold
GFZ German Research Centre for Geosciences + Humboldt-Universität zu Berlin; GFZ German Research Centre for Geosciences; GFZ German Research Centre for Geosciences; GFZ German Research Centre for Geosciences; Scientific Computing Center, Karlsruhe Institute of Technology; GFZ German Research Centre for Geosciences
Poster
main
https://github.com/GFZ/weedsgalore
https://openaccess.thecvf.com/content/WACV2025/html/Celikkan_WeedsGalore_A_Multispectral_and_Multitemporal_UAV-Based_Dataset_for_Crop_and_WACV_2025_paper.html
1
2502.13103
WeedsGalore: A Multispectral and Multitemporal UAV-Based Dataset for Crop and Weed Segmentation in Agricultural Maize Fields Weeds are one of the major reasons for crop yield loss but current weeding practices fail to manage weeds in an efficient and targeted manner. Effective weed management is especially important fo...
[ -0.05496544390916824, 0.015507061965763569, -0.0495494082570076, -0.013841996900737286, -0.0018137316219508648, -0.026037225499749184, -0.001781711122021079, 0.0551484189927578, -0.002785781864076853, 0.03575315326452255, -0.045011647045612335, -0.020328432321548462, 0.015470466576516628, ...
wacv_2025_10d4e4c35c
10d4e4c35c
wacv
2,025
Weight Copy and Low-Rank Adaptation for Few-Shot Distillation of Vision Transformers
Few-shot knowledge distillation recently emerged as a viable approach to harness the knowledge of large-scale pre-trained models using limited data and computational resources. In this paper we propose a novel few-shot feature distillation approach for vision transformers. Our approach is based on two key steps. Levera...
Diana-Nicoleta Grigore; Mariana-Iuliana Georgescu; Jon Alvarez Justo; Tor Johansen; Andreea Iuliana Ionescu; Radu Tudor Ionescu
University of Bucharest, Romania; University of Bucharest, Romania; Norwegian University of Science and Technology, Norway; Norwegian University of Science and Technology, Norway; University of Medicine and Pharmacy “Carol Davila”, Romania; University of Bucharest, Romania
Poster
main
https://github.com/dianagrigore/WeCoLoRA
https://openaccess.thecvf.com/content/WACV2025/html/Grigore_Weight_Copy_and_Low-Rank_Adaptation_for_Few-Shot_Distillation_of_Vision_WACV_2025_paper.html
1
2404.09326
Weight Copy and Low-Rank Adaptation for Few-Shot Distillation of Vision Transformers Few-shot knowledge distillation recently emerged as a viable approach to harness the knowledge of large-scale pre-trained models using limited data and computational resources. In this paper we propose a novel few-shot feature distilla...
[ -0.03638426214456558, -0.03640242666006088, 0.008455753326416016, 0.03996274247765541, 0.0035693994723260403, -0.027338149026036263, -0.001968619180843234, 0.045521195977926254, 0.015994179993867874, 0.018637169152498245, -0.01475896779447794, -0.0190731268376112, -0.013532837852835655, 0....
wacv_2025_a60d39bcc2
a60d39bcc2
wacv
2,025
When Cars Meet Drones: Hyperbolic Federated Learning for Source-Free Domain Adaptation in Adverse Weather
In Federated Learning (FL) multiple clients collaboratively train a global model without sharing private data. In semantic segmentation the Federated source Free Domain Adaptation (FFREEDA) setting is of particular interest where clients undergo unsupervised training after supervised pretraining at the server side. Whi...
Giulia Rizzoli; Matteo Caligiuri; Donald Shenaj; Francesco Barbato; Pietro Zanuttigh
University of Padova, Italy; University of Padova, Italy; ; ;
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Rizzoli_When_Cars_Meet_Drones_Hyperbolic_Federated_Learning_for_Source-Free_Domain_WACV_2025_paper.html
1
2403.13762
When Cars Meet Drones: Hyperbolic Federated Learning for Source-Free Domain Adaptation in Adverse Weather In Federated Learning (FL) multiple clients collaboratively train a global model without sharing private data. In semantic segmentation the Federated source Free Domain Adaptation (FFREEDA) setting is of particular...
[ -0.059074945747852325, -0.04046229273080826, -0.01830918714404106, -0.01566074602305889, -0.01635044440627098, -0.03260893002152443, 0.01735280640423298, 0.015752706676721573, 0.020267929881811142, 0.042558975517749786, -0.014906676486134529, -0.031873252242803574, 0.011007582768797874, 0....
wacv_2025_069dd94406
069dd94406
wacv
2,025
When Visual State Space Model Meets Backdoor Attacks
The recently proposed Visual State Space Model (VMamba) operating on the principle of state space mechanisms (SSM) processes images as a sequence of patches and outperforms Vision Transformers (ViT) in several computer vision tasks. Given their substantial design differences from CNNs and ViT it is crucial to investiga...
Sankalp Nagaonkar; Achyut Mani Tripathi; Ashish Mishra
IIT Dharwad; IIT Dharwad; HPE lab, Bangalore
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Nagaonkar_When_Visual_State_Space_Model_Meets_Backdoor_Attacks_WACV_2025_paper.html
0
When Visual State Space Model Meets Backdoor Attacks The recently proposed Visual State Space Model (VMamba) operating on the principle of state space mechanisms (SSM) processes images as a sequence of patches and outperforms Vision Transformers (ViT) in several computer vision tasks. Given their substantial design dif...
[ -0.006665025372058153, -0.002678754972293973, -0.045352380722761154, 0.021146822720766068, -0.010252432897686958, 0.026999961584806442, 0.026414647698402405, 0.03674260526895523, 0.03864959627389908, 0.009426385164260864, 0.0049043637700378895, 0.007599639240652323, -0.02596150152385235, -...
wacv_2025_3d243da6b1
3d243da6b1
wacv
2,025
Which Transformer to Favor: A Comparative Analysis of Efficiency in Vision Transformers
Self-attention in Transformers comes with a high computational cost because of their quadratic computational complexity but their effectiveness in addressing problems in language and vision has sparked extensive research aimed at enhancing their efficiency. However diverse experimental conditions spanning multiple inpu...
Tobias Christian Nauen; Sebastian Palacio; Federico Raue; Andreas Dengel
University of Kaiserslautern-Landau + German Research Center for Artificial Intelligence (DFKI); ABB AG; German Research Center for Artificial Intelligence (DFKI); University of Kaiserslautern-Landau + German Research Center for Artificial Intelligence (DFKI)
Poster
main
https://github.com/tobna/WhatTransformerToFavor
https://openaccess.thecvf.com/content/WACV2025/html/Nauen_Which_Transformer_to_Favor_A_Comparative_Analysis_of_Efficiency_in_WACV_2025_paper.html
2
2308.09372
Which Transformer to Favor: A Comparative Analysis of Efficiency in Vision Transformers Self-attention in Transformers comes with a high computational cost because of their quadratic computational complexity but their effectiveness in addressing problems in language and vision has sparked extensive research aimed at en...
[ -0.03176973760128021, -0.027455100789666176, -0.040159307420253754, 0.01003060769289732, 0.0023301802575588226, 0.010556108318269253, 0.006347494199872017, 0.017977651208639145, 0.032912932336330414, 0.005310321692377329, -0.03051591105759144, -0.007665854878723621, -0.03601061925292015, 0...
wacv_2025_82ee77b5a7
82ee77b5a7
wacv
2,025
Who Brings the Frisbee: Probing Hidden Hallucination Factors in Large Vision-Language Model via Causality Analysis
Recent advancements in large vision-language models (LVLM) have significantly enhanced their ability to comprehend visual inputs alongside natural language. However a major challenge in their real-world application is hallucination where LVLMs generate non-existent visual elements eroding user trust. The underlying mec...
Po-Hsuan Huang; Jeng-Lin Li; Chin-Po Chen; Ming-Ching Chang; Wei-Chao Chen
;;;;
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Huang_Who_Brings_the_Frisbee_Probing_Hidden_Hallucination_Factors_in_Large_WACV_2025_paper.html
0
2412.02946
Who Brings the Frisbee: Probing Hidden Hallucination Factors in Large Vision-Language Model via Causality Analysis Recent advancements in large vision-language models (LVLM) have significantly enhanced their ability to comprehend visual inputs alongside natural language. However a major challenge in their real-world ap...
[ -0.04981710761785507, -0.013717091642320156, -0.031680166721343994, -0.021650860086083412, -0.02882510796189308, 0.021101810038089752, 0.0009717035572975874, 0.007915466092526913, -0.0037564148660749197, 0.011017597280442715, -0.05311140790581703, -0.008469091728329659, -0.02593344636261463,...
wacv_2025_f5f24f74f7
f5f24f74f7
wacv
2,025
WiGNet: Windowed Vision Graph Neural Network
In recent years Graph Neural Networks (GNNs) have demonstrated strong adaptability to various real-world challenges with architectures such as Vision GNN (ViG) achieving state-of-the-art performance in several computer vision tasks. However their practical applicability is hindered by the computational complexity of co...
Gabriele Spadaro; Marco Grangetto; Attilio Fiandrotti; Enzo Tartaglione; Jhony H. Giraldo
University of Turin, Italy + LTCI, T ´el´ecom Paris, Institut Polytechnique de Paris, France; University of Turin, Italy; University of Turin, Italy + LTCI, T ´el´ecom Paris, Institut Polytechnique de Paris, France; LTCI, T ´el´ecom Paris, Institut Polytechnique de Paris, France; LTCI, T ´el´ecom Paris, Institut Polyte...
Poster
main
https://github.com/EIDOSLAB/WiGNet
https://openaccess.thecvf.com/content/WACV2025/html/Spadaro_WiGNet_Windowed_Vision_Graph_Neural_Network_WACV_2025_paper.html
1
2410.00807
WiGNet: Windowed Vision Graph Neural Network In recent years Graph Neural Networks (GNNs) have demonstrated strong adaptability to various real-world challenges with architectures such as Vision GNN (ViG) achieving state-of-the-art performance in several computer vision tasks. However their practical applicability is h...
[ -0.05280767008662224, -0.01856258325278759, -0.02188861183822155, -0.000238651831750758, -0.026478158310055733, -0.021163946017622948, 0.00018900528084486723, -0.012579448521137238, 0.0383886843919754, -0.0013030041009187698, -0.023096386343240738, -0.0004058823105879128, -0.0079155759885907...
wacv_2025_925193e0ad
925193e0ad
wacv
2,025
XPose: Towards Extreme Low Light Hand Pose Estimation
Recent advances in deep learning have enabled considerable strides in hand pose estimation in well-lit conditions. However to the best of our knowledge there is no existing method for hand pose estimation from RGB images captured in low light conditions. This task is highly challenging due to the overwhelming amount of...
Green Rosh; Meghana Shankar; Prateek Kukreja; Anmol Namdev; Pawan Prasad B H
Samsung R&D Institute India, Bangalore; Samsung R&D Institute India, Bangalore; Samsung R&D Institute India, Bangalore; Samsung R&D Institute India, Bangalore; Samsung R&D Institute India, Bangalore
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Rosh_XPose_Towards_Extreme_Low_Light_Hand_Pose_Estimation_WACV_2025_paper.html
0
XPose: Towards Extreme Low Light Hand Pose Estimation Recent advances in deep learning have enabled considerable strides in hand pose estimation in well-lit conditions. However to the best of our knowledge there is no existing method for hand pose estimation from RGB images captured in low light conditions. This task i...
[ -0.03570122271776199, -0.020050127059221268, -0.04984356090426445, 0.008370882831513882, -0.020504571497440338, 0.01640547625720501, 0.03402886539697647, 0.03424699977040291, 0.017087142914533615, -0.0044353860430419445, -0.0290299654006958, -0.03126583620905876, 0.013442492112517357, -0.0...
wacv_2025_ad2f97206e
ad2f97206e
wacv
2,025
XR-MBT: Multi-Modal Full Body Tracking for XR through Self-Supervision with Learned Depth Point Cloud Registration
Tracking the full body motions of users in XR (AR/VR) devices is a fundamental challenge to bring a sense of authentic social presence. Due to the absence of dedicated leg sensors currently available body tracking methods adopt a synthesis approach to generate plausible motions given a 3-point signal from the head and ...
Denys Rozumnyi; Nadine Bertsch; Othman Sbai; Filippo Arcadu; Yuhua Chen; Artsiom Sanakoyeu; Manoj Kumar; Catherine Herold; Robin Kips
Meta Reality Labs Zurich; Meta Reality Labs Zurich; Meta Reality Labs Zurich; Meta Reality Labs Zurich; Meta Reality Labs Zurich; Meta Reality Labs Zurich; Meta Reality Labs Zurich; Meta Reality Labs Zurich; Meta Reality Labs Zurich
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Rozumnyi_XR-MBT_Multi-Modal_Full_Body_Tracking_for_XR_through_Self-Supervision_with_WACV_2025_paper.html
3
XR-MBT: Multi-Modal Full Body Tracking for XR through Self-Supervision with Learned Depth Point Cloud Registration Tracking the full body motions of users in XR (AR/VR) devices is a fundamental challenge to bring a sense of authentic social presence. Due to the absence of dedicated leg sensors currently available body ...
[ -0.01876768469810486, -0.03447239473462105, -0.033896926790475845, 0.0198072399944067, -0.023334303870797157, 0.0175889041274786, 0.032393284142017365, -0.0018957961583510041, 0.007884841412305832, -0.014247476123273373, -0.027418270707130432, -0.03838929161429405, 0.025654738768935204, 0....
wacv_2025_39b4836189
39b4836189
wacv
2,025
ZAHA: Introducing the Level of Facade Generalization and the Large-Scale Point Cloud Facade Semantic Segmentation Benchmark Dataset
Facade semantic segmentation is a long-standing challenge in photogrammetry and computer vision. Although the last decades have witnessed the influx of facade segmentation methods there is a lack of comprehensive facade classes and data covering the architectural variability. In ZAHA we introduce Level of Facade Genera...
Olaf Wysocki; Yue Tan; Thomas Froech; Yan Xia; Magdalena Wysocki; Ludwig Hoegner; Daniel Cremers; Christoph Holst
Technical University of Munich; Technical University of Munich; Technical University of Munich; Technical University of Munich+Munich Center for Machine Learning (MCML); Technical University of Munich+Munich Center for Machine Learning (MCML); Munich University of Applied Sciences; Technical University of Munich+Munich...
Poster
main
https://github.com/OloOcki/zaha
https://openaccess.thecvf.com/content/WACV2025/html/Wysocki_ZAHA_Introducing_the_Level_of_Facade_Generalization_and_the_Large-Scale_WACV_2025_paper.html
1
2411.04865
ZAHA: Introducing the Level of Facade Generalization and the Large-Scale Point Cloud Facade Semantic Segmentation Benchmark Dataset Facade semantic segmentation is a long-standing challenge in photogrammetry and computer vision. Although the last decades have witnessed the influx of facade segmentation methods there is...
[ -0.048206303268671036, -0.04444996640086174, -0.006269765552133322, -0.04050949588418007, 0.003737921826541424, 0.02603655867278576, 0.025336848571896553, -0.00643088249489665, -0.021451620385050774, -0.017953071743249893, -0.02850395441055298, 0.028982704505324364, 0.0030405139550566673, ...
wacv_2025_d8e4afe9e5
d8e4afe9e5
wacv
2,025
Zero-Shot Class Unlearning in CLIP with Synthetic Samples
Machine unlearning is a crucial area of research. It is driven by the need to remove sensitive information from models to safeguard individuals' right to be forgotten under rigorous regulations such as GDPR. In this work we focus on unlearning within CLIP a dual vision-language encoder model trained on a massive datase...
Alexey Kravets; Vinay Namboodiri
University of Bath, Bath, UK; University of Bath, Bath, UK
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Kravets_Zero-Shot_Class_Unlearning_in_CLIP_with_Synthetic_Samples_WACV_2025_paper.html
5
2407.07485
Zero-Shot Class Unlearning in CLIP with Synthetic Samples Machine unlearning is a crucial area of research. It is driven by the need to remove sensitive information from models to safeguard individuals' right to be forgotten under rigorous regulations such as GDPR. In this work we focus on unlearning within CLIP a dual...
[ -0.07141052931547165, -0.01052080187946558, -0.006588359829038382, 0.016795875504612923, 0.0006142979254946113, -0.043168019503355026, 0.03744470328092575, 0.04915318638086319, -0.009099324233829975, 0.005933732260018587, -0.02373492904007435, -0.011586910113692284, -0.005386650562286377, ...
wacv_2025_b94466bbb0
b94466bbb0
wacv
2,025
Zero-Shot Detection of Out-of-Context Objects using Foundation Models
We address the problem of detecting out-of-context (OOC) objects in a scene. Given an image we aim to detect whether the image has objects that are not present in their usual context and localize such OOC objects. Existing approaches for OOC detection rely on defining the common context in terms of the manually constru...
Anirban Roy; Adam Cobb; Ramneet Kaur; Sumit Jha; Nathaniel Bastian; Alexander Berenbeim; Robert Thomson; Iain Cruickshank; Alvaro Velasquez; Susmit Jha
SRI; SRI; SRI; Florida International University; United States Military Academy; United States Military Academy; United States Military Academy; United States Military Academy; DARPA; SRI
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Roy_Zero-Shot_Detection_of_Out-of-Context_Objects_using_Foundation_Models_WACV_2025_paper.html
0
Zero-Shot Detection of Out-of-Context Objects using Foundation Models We address the problem of detecting out-of-context (OOC) objects in a scene. Given an image we aim to detect whether the image has objects that are not present in their usual context and localize such OOC objects. Existing approaches for OOC detectio...
[ -0.02575812302529812, -0.01936362497508526, -0.06787175685167313, 0.01780552789568901, 0.0061153024435043335, -0.03559304401278496, 0.00021375977667048573, 0.03728623315691948, 0.03636758774518967, 0.0185890793800354, -0.0473012812435627, -0.03209858387708664, 0.0000980846380116418, 0.0322...
wacv_2025_030c6ece00
030c6ece00
wacv
2,025
ZeroComp: Zero-Shot Object Compositing from Image Intrinsics via Diffusion
We present ZeroComp an effective zero-shot 3D object compositing approach that does not require paired composite-scene images during training. Our method leverages ControlNet to condition from intrinsic images and combines it with a Stable Diffusion model to utilize its scene priors together operating as an effective r...
Zitian Zhang; Frédéric Fortier-Chouinard; Mathieu Garon; Anand Bhattad; Jean-François Lalonde
;;;;
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Zhang_ZeroComp_Zero-Shot_Object_Compositing_from_Image_Intrinsics_via_Diffusion_WACV_2025_paper.html
2
2410.08168
ZeroComp: Zero-Shot Object Compositing from Image Intrinsics via Diffusion We present ZeroComp an effective zero-shot 3D object compositing approach that does not require paired composite-scene images during training. Our method leverages ControlNet to condition from intrinsic images and combines it with a Stable Diffu...
[ -0.0443701334297657, -0.006703072227537632, -0.025001637637615204, 0.03537173941731453, -0.0018220834899693727, 0.0035435750614851713, 0.025404004380106926, 0.014631535857915878, -0.015317389741539955, 0.02993978001177311, -0.028878994286060333, -0.024654138833284378, 0.0229166429489851, -...
wacv_2025_24f5fc14f4
24f5fc14f4
wacv
2,025
eLIR-Net: An Efficient AI Solution for Image Retouching
Picture quality serves as a primary differentiator for prominent display panel manufacturers. AI-based solutions have made remarkable progress in delivering expert-level image color remastering operations. However their demand on intensive computation resources heavily impedes the on-device usage in industries where sp...
Tingting Zhao; Chenguang Liu; Kamal Jnawali; Chang Su
Digital Media Solutions Lab, Samsung Research America, Irvine, CA, USA; Digital Media Solutions Lab, Samsung Research America, Irvine, CA, USA; Digital Media Solutions Lab, Samsung Research America, Irvine, CA, USA; Digital Media Solutions Lab, Samsung Research America, Irvine, CA, USA
Poster
main
https://openaccess.thecvf.com/content/WACV2025/html/Zhao_eLIR-Net_An_Efficient_AI_Solution_for_Image_Retouching_WACV_2025_paper.html
0
eLIR-Net: An Efficient AI Solution for Image Retouching Picture quality serves as a primary differentiator for prominent display panel manufacturers. AI-based solutions have made remarkable progress in delivering expert-level image color remastering operations. However their demand on intensive computation resources he...
[ -0.03272811323404312, -0.016209421679377556, -0.009714736603200436, 0.031909454613924026, -0.018192391842603683, 0.01926574297249317, -0.0029812781140208244, 0.05759711191058159, -0.002109180437400937, 0.041042033582925797, -0.0051302541978657246, -0.016182132065296173, -0.03367411717772484,...
wacv_2025_0ff9bc196e
0ff9bc196e
wacv
2,025
uLayout: Unified Room Layout Estimation for Perspective and Panoramic Images
We present uLayout a unified model for estimating room layout geometries from both perspective and panoramic images whereas traditional solutions require different model designs for each image type. The key idea of our solution is to unify both domains into the equirectangular projection particularly allocating perspec...
Jonathan Lee; Bolivar E Solarte; Chin-Hsuan Wu; Jin-Cheng Jhang; Fu-En Wang; Yi-Hsuan Tsai; Min Sun
National Tsing Hua University, Taiwan; National Tsing Hua University, Taiwan; National Tsing Hua University, Taiwan; National Tsing Hua University, Taiwan; National Tsing Hua University, Taiwan; Atmanity Inc.; National Tsing Hua University, Taiwan
Poster
main
https://github.com/JonathanLee112/uLayout
https://openaccess.thecvf.com/content/WACV2025/html/Lee_uLayout_Unified_Room_Layout_Estimation_for_Perspective_and_Panoramic_Images_WACV_2025_paper.html
0
uLayout: Unified Room Layout Estimation for Perspective and Panoramic Images We present uLayout a unified model for estimating room layout geometries from both perspective and panoramic images whereas traditional solutions require different model designs for each image type. The key idea of our solution is to unify bot...
[ -0.08811119943857193, -0.013433552347123623, -0.04488007351756096, -0.0072275917045772076, -0.007115570828318596, 0.029232988134026527, -0.009606916457414627, 0.0019749293569475412, 0.02509269304573536, 0.030595162883400917, -0.0019805303309112787, 0.029394296929240227, -0.005309793166816235...