cleaned up scripts

Code/Inference.py

@@ -120,7 +120,6 @@ def DLCInference(Crop,dlc_liveObj,CropSize):
 
 def VisualizeAll(frame, box, DLCPredict2D,MeanConfidence,ScaleBBox, imsize):
     """Visualize all stuff"""
-    # colourList = [(255,255,0),(255,0 ,255),(128,0,128),(203,192,255),(0, 255, 255),(255, 0 , 0 ),(63,133,205),(0,255,0),(0,0,255)]
     colourList = [(0,255,255),(255,0 ,255),(128,0,128),(255,192,203),(255, 255, 0),(0, 0 , 255 ),(205,133,63),(0,255,0),(255,0,0)]
 
     ##Order: Lshoulder, Rshoulder, topKeel,botKeel,Tail,Beak,Nose,Leye,Reye
@@ -131,23 +130,8 @@ def VisualizeAll(frame, box, DLCPredict2D,MeanConfidence,ScaleBBox, imsize):
         cv2.circle(frame,roundPoint,1,colourList[x], 5)
         PlotPoints.append(roundPoint)
 
-
-    ##change box to XYWH to scale down
-    # bbox = [box[0],box[1],box[2]-box[0],box[3]-box[1]]
-    # ScaleWidth = (bbox[2]/ScaleBBox)/2
-    # ScaleHeight = (bbox[3]/ScaleBBox)/2
-    # ###based on ScaleBBox, scale back down bounding box for plotting
-    # x1 = round(bbox[0]+ScaleWidth) if round(bbox[0]+ScaleWidth)>0 else 0
-    # y1 = round(bbox[1]+ScaleHeight)if round(bbox[1]+ScaleHeight)>0 else 0
-    # x2 = round(bbox[0]+bbox[2]-ScaleWidth) if round(bbox[0]+bbox[2]-ScaleWidth) < imsize[0] else imsize[0]
-    # y2 = round(bbox[1]+bbox[3]-ScaleHeight)if round(bbox[1]+bbox[3]-ScaleHeight) < imsize[1] else imsize[1]
-    # box = [x1,y1,x2,y2]
     cv2.rectangle(frame,(round(box[0]),round(box[1])),(round(box[2]),round(box[3])),[0,0,255],3)
 
-    #plot mean confidence
-    # import ipdb;ipdb.set_trace()
-    # font = cv2.FONT_HERSHEY_SIMPLEX
-    # cv2.putText(frame,str(round(MeanConfidence,3)),(round(box[0]),round(box[1])),font,2,[255,0,0],2)
 
     return frame, PlotPoints
 
@@ -201,75 +185,13 @@ def Inference(frame,predictor,dlc_liveObj,ScaleBBox=1,Dilate=5,DLCThreshold=0.3)
         BirdCrop = Crop[y1:y2,x1:x2] #bbox is XYWH
 
         DLCPredict2D= DLCInference(BirdCrop,dlc_liveObj,CropSize)
-        MeanConfidence = DLCPredict2D[:,2].mean()
 
-        if MeanConfidence > DLCThreshold: #if mean keypoint confidence is higher than this threshold, consider bird
-            bbox.append(MeanConfidence)
-            frame, PlotPoints = VisualizeAll(frame, bbox, DLCPredict2D,MeanConfidence,ScaleBBox,imsize)
+        frame, PlotPoints = VisualizeAll(frame, bbox, DLCPredict2D,MeanConfidence,ScaleBBox,imsize)
 
     if BirdBBox.shape[0] == 0:
         DLCPredict2D= DLCInference(InferFrame,dlc_liveObj,CropSize)
-        MeanConfidence = DLCPredict2D[:,2].mean()
         bbox = [0,0,0,0]
-        if MeanConfidence > DLCThreshold: #if mean keypoint confidence is higher than this threshold, consider bird
-            frame, PlotPoints = VisualizeAll(frame, bbox, DLCPredict2D,MeanConfidence,ScaleBBox,imsize)
+        frame, PlotPoints = VisualizeAll(frame, bbox, DLCPredict2D,MeanConfidence,ScaleBBox,imsize)
 
     return frame
 
-
-
-
-if __name__ == "__main__":
-    # VidPath = "/media/alexchan/Extreme SSD/WorkDir/Pigeon3DTrack/ManualClickTrials/Videos/Cam1_C0008.MP4_Trimmed.mp4"
-    VidPath = "/media/alexchan/Extreme SSD/SampleDatasets/PigeonsEverywhere/2022-12-01_Pilot_bottom_control.mp4"
-
-    OutDir = "/media/alexchan/Extreme SSD/WorkDir/Pigeon3DTrack/OutdoorTracking/Final2D"
-
-    ExportModelPath = "/media/alexchan/Extreme SSD/WorkDir/Pigeon3DTrack/Weights/DLC_Weights/N6000_DLC_Mask/exported-models/DLC_DLC_Segmented_resnet_50_iteration-0_shuffle-1"
-    # ExportModelPath = "/media/alexchan/Extreme SSD/WorkDir/Pigeon3DTrack/Weights/DLC_Weights/DLC_PigeonSuperModel_imgaug_efficientnet-b0_iteration-0_shuffle-2"
-
-    if not os.path.isdir(ExportModelPath):
-        #need to export model first
-        # DLC_Config = "/media/alexchan/Extreme SSD/SampleDatasets/ImageTrainingData/N5000/DLC_Seg_Aug/config.yaml"
-
-        dlc.export_model(DLC_Config)
-    else:
-        print("model already exported!")
-
-    CropSize = (320,320)
-    # WeightPath = "/home/alexchan/Documents/Pigeon3DTrack/Data/YOLO_Weights/yolov8m.pt"
-    # VidPath= "/media/alexchan/Extreme SSD/PigeonOutdoors/28032023/Cam1_C0004.MP4"
-    # import ipdb;ipdb.set_trace()
-    # YOLOModel = YOLO(YOLOPath)
-    device = "cuda"
-    # device = "cpu"
-
-    ###Detectron:
-    cfg = get_cfg()
-    # add project-specific config (e.g., TensorMask) here if you're not running a model in detectron2's core library
-    cfg.merge_from_file(model_zoo.get_config_file("COCO-InstanceSegmentation/mask_rcnn_R_101_FPN_3x.yaml"))
-    cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.7  # set threshold for this model
-    # Find a model from detectron2's model zoo. You can use the https://dl.fbaipublicfiles... url as well
-    cfg.MODEL.WEIGHTS = model_zoo.get_checkpoint_url("COCO-InstanceSegmentation/mask_rcnn_R_101_FPN_3x.yaml")
-    predictor = DefaultPredictor(cfg)
-
-
-    ##DLC:
-    dlc_proc = Processor()
-    dlc_liveObj = DLCLive(ExportModelPath, processor=dlc_proc)
-    
-    RunInference(predictor,dlc_liveObj, device,OutDir,
-                 VidPath,CropSize,startFrame=0,
-                 TotalFrames =900, ScaleBBox=1.3,
-                 DLCThreshold = 0, Dilate=5)
-    
-
-
-    ###DLC tests
-    # DLC_Config = "/media/alexchan/Extreme SSD/SampleDatasets/ImageTrainingData/N5000/DLC/config.yaml"
-    # Video = "/media/alexchan/Extreme SSD/SampleDatasets/ImageTrainingData/N5000/DLC/videos/Video1.mp4"
-    # dlc.analyze_videos(DLC_Config,[Video])
-    # dlc.export_model(DLC_Config)
-
-    # import pandas as pd
-    # pd.read_hdf("/media/alexchan/Extreme SSD/SampleDatasets/ImageTrainingData/N5000/DLC/videos/Video1DLC_resnet50_DLC20230401-173058shuffle1_3000.h5")

app.py

@@ -1,6 +1,6 @@
 import sys
 import subprocess
-# subprocess.check_call([sys.executable, '-m', 'pip', 'install','git+https://github.com/facebookresearch/detectron2.git'])
+subprocess.check_call([sys.executable, '-m', 'pip', 'install','git+https://github.com/facebookresearch/detectron2.git'])
 
 import numpy as np
 import gradio as gr