Learning to Forecast and Refine Residual Motion for Image-to-Video Generation.
Long ZhaoXi PengYu TianMubbasir KapadiaDimitris N. MetaxasPublished in: CoRR (2018)
Keyphrases
- image features
- image frames
- image data
- visual cues
- learning algorithm
- static images
- key frames
- object motion
- input image
- image classification
- visual data
- single frame
- space time
- image content
- image segmentation
- image retrieval
- low level
- normalized correlation
- temporal continuity
- camera movement
- video scene
- multiscale
- single image
- image representation
- high resolution
- visual perception
- position and orientation
- vector field
- bounding box
- video retrieval
- motion segmentation
- human motion
- motion model
- video frames