447 citations · 480 across the 6 of their papers we have counts for
1 paper · 2 filters
Bo Li, Huahui Chen, Yucheng Chen +2
Action recognition from well-segmented 3D skeleton video has been intensively studied. However, due to the difficulty in representing the 3D skeleton video and the lack of training…