Generalized Label-Efficient 3D Scene Parsing via Hierarchical Feature Aligned Pre-Training and Region-Aware Fine-tuning.
Kangcheng LiuYong-Jin LiuKai TangMing LiuBaoquan ChenPublished in: CoRR (2023)
Keyphrases
- d scene
- fine tuning
- depth map
- single image
- optical flow
- scene structure
- planar surfaces
- scene understanding
- image features
- fine tuned
- camera parameters
- coordinate frame
- viable alternative
- flow estimation
- scene flow
- image based rendering
- image processing
- viewpoint
- view synthesis
- coarse to fine
- least squares
- semi supervised
- camera viewpoint
- similarity measure