CAM-Convs: Camera-Aware Multi-Scale Convolutions for Single-View Depth.
José M. FácilBenjamin UmmenhoferHuizhong ZhouLuis MontesanoThomas BroxJavier CiveraPublished in: CVPR (2019)
Keyphrases
- single view
- multi view
- multiscale
- depth map
- calibrated cameras
- vanishing points
- multiple cameras
- camera parameters
- camera pose
- depth estimation
- multiple views
- camera calibration
- time of flight
- perspective projection
- field of view
- depth information
- motion stereo
- co training
- single camera
- polyhedral objects
- camera motion
- scale space
- d scene
- ground plane
- ego motion
- edge detection
- repetitive structures
- focal length
- bundle adjustment
- vision system
- video camera
- stereo matching
- multiple viewpoints
- single view reconstruction
- depth images
- stereo vision
- structure from motion
- semi supervised
- three dimensional
- position and orientation
- multi camera
- surface reconstruction
- photometric stereo
- shape from shading
- feature points
- light directions
- viewpoint
- computer vision