DS-MVSNet: Unsupervised Multi-view Stereo via Depth Synthesis.
Jingliang LiZhengda LuYiqun WangYing WangJun XiaoPublished in: CoRR (2022)
Keyphrases
- multi view stereo
- depth map
- depth estimation
- multi view
- depth information
- stereo vision
- stereo matching
- d scene
- stereo algorithms
- three dimensional
- view synthesis
- high quality
- semi supervised
- real scenes
- stereo pair
- depth images
- image sequences
- point cloud
- bundle adjustment
- input image
- stereo reconstruction
- viewpoint
- depth data
- structure from motion
- super resolution
- camera parameters
- surface reconstruction
- scene understanding
- stereo images
- ray tracing
- low resolution
- urban environments
- high fidelity
- computer vision
- scene reconstruction
- vision system
- stereo camera
- dynamic scenes
- high resolution
- active learning
- camera calibration
- least squares