V-Express: Conditional Dropout for Progressive Training of Portrait Video Generation.
Cong WangKuan TianJun ZhangYonghang GuanFeng LuoFei ShenZhiwei JiangQing GuXiao HanWei YangPublished in: CoRR (2024)
Keyphrases
- video data
- video content
- multimedia
- training set
- training process
- video analysis
- video sequences
- real time
- video streams
- digital video
- supervised learning
- spatial and temporal
- real time video
- video clips
- dynamic scenes
- training algorithm
- random field model
- video frames
- key frames
- space time
- image compression
- multiscale
- feature selection