Residual attention-based LSTM for video captioning.
Xiangpeng LiZhilong ZhouLijiang ChenLianli GaoPublished in: World Wide Web (2019)
Keyphrases
- video sequences
- video data
- video streams
- real time
- video content
- spatial and temporal
- video database
- spatial temporal
- quality metrics
- video images
- multimedia
- video frames
- space time
- surveillance videos
- compressed video
- recurrent neural networks
- digital video
- residual error
- visual attention
- focus of attention
- visual data
- video analysis
- video surveillance
- key frames
- temporal information
- human activities
- multi modal
- high level
- metadata