1 paper
Duncan Zauss, Sven Kreiss, Alexandre Alahi
We present a fast bottom-up method that jointly detects over 100 keypoints on humans or objects, also referred to as human/object pose estimation. We model all keypoints belonging…