MAPLE: Masked Pseudo-Labeling autoEncoder for Semi-supervised Point Cloud Action Recognition.
Xiaodong ChenWu LiuXinchen LiuYongdong ZhangJungong HanTao MeiPublished in: CoRR (2022)
Keyphrases
- action recognition
- point cloud
- semi supervised
- active learning
- structure from motion
- human actions
- surface reconstruction
- semi supervised learning
- computer vision
- activity recognition
- point sets
- labeled data
- recognizing human actions
- supervised learning
- action classification
- multi view
- body parts
- laser scanner
- stereo camera
- recognition of human actions
- image segmentation
- human activities
- pairwise
- learning algorithm
- three dimensional
- feature vectors
- action recognition in videos