slightly tweaked mocapnet alt container

animation/MocapNET-kasisnu/src/python/datasetAnnotations/annotate.py

@@ -0,0 +1,43 @@
+
+# import required module 
+import sys
+import cv2 
+import numpy as np
+from datetime import datetime 
+
+
+pathToDataset="/home/ammar/Documents/Datasets/ai_challenger"
+filename = "%s/train/annotation.json" % pathToDataset
+
+
+import json
+with open(filename) as jsonData:
+          data = json.load(jsonData)
+
+print (len(data))
+print (data.keys())
+if ( len(data['images']) != len(data['annotations'])  ):
+   print("Mismatching image/annotation data")
+   sys.exit(0)
+
+
+for imageID in range(0,len(data['images'])):
+   imageData=data['images'][imageID]
+   imageAnnotations=data['annotations'][imageID]
+   print(imageData)
+   loadImage=cv2.imread("%s/train/images/%s"%(pathToDataset,imageData['file_name']),-1)
+   
+   #Draw Bounding Box   
+   cv2.rectangle(loadImage, (imageAnnotations['bbox'][0],imageAnnotations['bbox'][1]),(imageAnnotations['bbox'][2],imageAnnotations['bbox'][3]), (255,255,255), 5)
+
+   radius = 5
+   color = (123,123,123)
+   
+   for keypointID in range(0,int( len(imageAnnotations['keypoints'])/3 ) ):
+      cv2.circle(loadImage, (imageAnnotations['keypoints'][keypointID*3+0],imageAnnotations['keypoints'][keypointID*3+1]) , radius, color, -1)
+
+   cv2.imshow("Annotations", loadImage)
+   print(imageAnnotations)
+
+   cv2.waitKey(1000)  
+