No Time to Train: Empowering Non-Parametric Networks for Few-shot 3D Scene Segmentation.
Xiangyang ZhuRenrui ZhangBowei HeZiyu GuoJiaming LiuHan XiaoChaoyou FuHao DongPeng GaoPublished in: CoRR (2024)
Keyphrases
- d scene
- rigidly moving
- scene representation
- single image
- image based rendering
- depth map
- scene understanding
- camera parameters
- image segmentation
- optical flow
- segmentation method
- segmentation algorithm
- planar surfaces
- camera viewpoint
- scene flow
- scene geometry
- video sequences
- flow estimation
- indoor scenes
- optical flow estimation
- object segmentation
- camera positions
- viewpoint
- view synthesis
- computational complexity
- motion segmentation
- image data