VidTr: Video Transformer Without Convolutions.
Yanyi ZhangXinyu LiChunhui LiuBing ShuaiYi ZhuBiagio BrattoliHao ChenIvan MarsicJoseph TighePublished in: ICCV (2021)
Keyphrases
- video content
- video data
- multimedia
- video sequences
- video analysis
- video streams
- video frames
- multimedia data
- video retrieval
- real time
- fuzzy logic
- compressed video
- video shots
- video clips
- video processing
- spatial and temporal
- space time
- video images
- digital video
- high voltage
- video segmentation
- video database
- dynamic scenes
- video surveillance
- key frames
- event detection
- multiscale
- signal processing