SigFormer: Sparse Signal-Guided Transformer for Multi-Modal Human Action Segmentation.
Qi LiuXinchen LiuKun LiuXiaoyan GuWu LiuPublished in: CoRR (2023)
Keyphrases
- multi modal
- human actions
- action recognition
- high dimensional
- spatio temporal
- human motion
- multi modality
- cross modal
- segmentation algorithm
- space time
- segmentation method
- image segmentation
- video sequences
- audio visual
- motion recognition
- human activities
- activity recognition
- medical images
- medical imaging
- bag of words
- image features
- computer vision
- uni modal
- multimedia