TwinFormer: Fine-to-Coarse Temporal Modeling for Long-Term Action Recognition.
Jiaming ZhouKun-Yu LinYukun QiuWei-Shi ZhengPublished in: IEEE Trans. Multim. (2024)
Keyphrases
- action recognition
- long term
- spatial temporal
- human actions
- bag of words
- human detection
- computer vision
- body parts
- activity recognition
- action classification
- coarse to fine
- motion history images
- spatio temporal
- bag of features
- video dataset
- human object interactions
- depth sensors
- recognition of human actions
- recognizing actions
- multiscale
- human pose
- mid level
- static images
- recognizing human actions
- independent subspace analysis
- space time interest points
- view invariant action recognition
- temporal resolution
- depth cameras
- machine learning
- temporal reasoning
- feature vectors
- temporal information
- spatial and temporal