pos = tracker.get_position()
# unpack the position object
startX = int(pos.left())
startY = int(pos.top())
endX = int(pos.right())
endY = int(pos.bottom())
# add the bounding box coordinates to the rectangles list
rects.append((startX, startY, endX, endY))
# check to see if we should look for new detections
if totalFrames % args["skip_frames"] == 0:
# detect people
results = detect_people(frame,
net,
ln,
personIdx=LABELS.index("person"))
for r in results:
startX = r[1][0]
startY = r[1][1]
endX = r[1][2]
endY = r[1][3]
r_cX = int((startX + endX) / 2.0)
r_cY = int((startY + endY) / 2.0)
r_c = (r_cX, r_cY)
new_tracker = True
if ret == False:
break
image = cv2.resize(image, (720, 640))
# cv2.namedWindow("output", cv2.WINDOW_NORMAL)
(H, W) = image.shape[:2]
ln = net.getLayerNames()
ln = [ln[i[0] - 1] for i in net.getUnconnectedOutLayers()]
blob = cv2.dnn.blobFromImage(image,
1/255.0,
(416, 416),
swapRB=True,
crop=False)
results = detect_people(image, net, ln,
personIdx=Labels.index("person"))
net.setInput(blob)
start = time.time()
layerOutputs = net.forward(ln)
end = time.time()
print("Time taken to predict the image: {:.6f}seconds".format(end-start))
boxes = []
confidences = []
classIDs = []
for output in layerOutputs:
for detection in output:
scores = detection[5:]
classID = np.argmax(scores)