Masked Motion Predictors are Strong 3D Action Representation Learners.
Yunyao MaoJiajun DengWengang ZhouYao FangWanli OuyangHouqiang LiPublished in: CoRR (2023)
Keyphrases
- recognizing human actions
- recognition of human actions
- action recognition
- view invariant
- learning process
- optical flow
- motion analysis
- image representation
- image sequences
- learning environment
- collaborative learning
- space time
- motion parameters
- human actions
- motion model
- e learning
- multi view
- motion estimation
- camera motion
- object shape
- decision trees
- motion capture
- learning scenarios
- language learning
- learning systems
- moving objects