Decouple Content and Motion for Conditional Image-to-Video Generation.
Cuifeng ShenYulu GanChen ChenXiongwei ZhuLele ChengJinzhi WangPublished in: CoRR (2023)
Keyphrases
- static images
- object motion
- image frames
- visual data
- key frames
- input image
- successive frames
- image content
- image classification
- multiscale
- camera movement
- visual cues
- video frames
- temporal continuity
- image features
- normalized correlation
- feature points
- video footage
- image data
- space time
- single image
- single frame
- camera motion
- input video
- layered representation
- image sequences
- image segmentation
- image retrieval
- motion estimation
- reference frame
- motion analysis
- moving objects
- multi frame
- visual concepts
- image motion
- video sequences
- multimedia data
- segmentation method
- optical flow
- high resolution
- visual features
- temporal domain
- low level
- video signals
- dynamic scenes
- motion trajectories
- position and orientation
- moving camera
- image collections