HaltingVT: Adaptive Token Halting Transformer for Efficient Video Recognition.
Qian WuRuoxuan CuiYuke LiHaoqi ZhuPublished in: ICASSP (2024)
Keyphrases
- recognition rate
- object recognition
- recognition accuracy
- spatio temporal
- video content
- human activities
- fault diagnosis
- feature extraction
- multimedia
- real time
- automatic recognition
- activity detection
- character recognition
- video images
- recognition algorithm
- video retrieval
- online video
- video frames
- space time
- image classification
- pattern recognition
- video sequences
- genetic algorithm
- neural network