Neighbor-Guided Pseudo-Label Generation and Refinement for Single-Frame Supervised Temporal Action Localization.
Guozhang LiDe ChengNannan WangJie LiXinbo GaoPublished in: IEEE Trans. Image Process. (2024)
Keyphrases
- single frame
- temporal redundancy
- super resolution
- multi frame
- image sequences
- multiple frames
- video sequences
- low resolution
- spatio temporal
- video frames
- high resolution video
- optical flow
- temporal coherence
- learning algorithm
- temporal information
- human actions
- three dimensional
- semi supervised
- surveillance videos
- machine learning
- supervised learning
- object detection
- space time
- high resolution
- motion analysis
- feature selection