Deep Multi-modal Representation Schemes for Federated 3D Human Action Recognition.
Athanasios PsaltisCharalampos Z. PatrikakisPetros DarasPublished in: ECCV Workshops (6) (2022)
Keyphrases
- multi modal
- action recognition
- static images
- human movements
- human activities
- human actions
- human detection
- activity recognition
- bag of words
- computer vision
- motion history images
- audio visual
- body parts
- human object interactions
- cross modal
- action classification
- multi modality
- recognition of human actions
- recognizing human actions
- motion capture data
- video search
- semantic concepts
- view invariant
- image annotation
- action detection
- high dimensional
- machine learning
- image registration
- higher level
- recognizing actions
- human motion
- action recognition in videos
- uni modal