Thanks. I was able to install and run the pre-trained model.
Question: (how to)
Is there a way to remove the background (bg=black) and only show the key point detections (skeletons)? I know the original openpose had a flag that controlled this feature. I will take a look at pose_detector.py draw_person_pose function...
Update: 1/13
One Solution was to modify the video capture image after pose detector was called:
https://pastebin.com/CC2dV6MS
person_pose_array, _ = pose_detector(img)
img = np.zeros(img.shape) #kkhatak - line of code added to make background black ###
res_img = draw_person_pose(img, person_pose_array)
res_img = cv2.addWeighted(img, 0.6, draw_person_pose(img, person_pose_array), 0.4, 0)
output_movie.write(np.uint8(res_img)) #kkhatak
Thanks. I was able to install and run the pre-trained model.
Question: (how to) Is there a way to remove the background (bg=black) and only show the key point detections (skeletons)? I know the original openpose had a flag that controlled this feature. I will take a look at pose_detector.py draw_person_pose function...
Update: 1/13 One Solution was to modify the video capture image after pose detector was called: https://pastebin.com/CC2dV6MS