VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking.
Limin WangBingkun HuangZhiyu ZhaoZhan TongYinan HeYi WangYali WangYu QiaoPublished in: CoRR (2023)
Keyphrases
- video data
- video sequences
- multimedia
- real time video
- video content
- video streams
- video database
- denoising
- video frames
- video clips
- video segmentation
- video analysis
- key frames
- event recognition
- digital video
- handwritten digits
- real time
- online video
- video retrieval
- video surveillance
- human visual system
- spatial and temporal
- low level
- multiresolution
- computer vision