SinFusion: Training Diffusion Models on a Single Image or Video.
Yaniv NikankinNiv HaimMichal IraniPublished in: CoRR (2022)
Keyphrases
- single image
- diffusion models
- multiple images
- shape recovery
- d scene
- light source
- diffusion model
- video frames
- video data
- video sequences
- shape from shading
- indoor scenes
- video content
- information diffusion
- intrinsic image decomposition
- super resolution
- video surveillance
- partial differential equations
- key frames
- image data
- social networks