LoLep: Single-View View Synthesis with Locally-Learned Planes and Self-Attention Occlusion Inference.
Cong WangYu-Ping WangDinesh ManochaPublished in: ICCV (2023)
Keyphrases
- view synthesis
- multi view
- single view
- vanishing points
- d objects
- three dimensional
- depth map
- multiple views
- multiple cameras
- photometric stereo
- semi supervised
- dynamic scenes
- camera calibration
- d scene
- stereo matching
- viewpoint
- object recognition
- camera parameters
- surface reconstruction
- range images
- pairwise
- depth estimation
- photorealistic
- stereo correspondence
- computer vision