Geometry meets semantics for semi-supervised monocular depth estimation.
Pierluigi Zama RamirezMatteo PoggiFabio TosiStefano MattocciaLuigi di StefanoPublished in: CoRR (2018)
Keyphrases
- depth estimation
- monocular images
- semi supervised
- multi view stereo
- depth cues
- depth map
- stereo vision
- multi view
- binocular vision
- binocular stereo
- stereo matching
- dynamic scenes
- depth information
- depth estimates
- three dimensional
- real scenes
- scene understanding
- stereo pair
- image sequences
- feature matching
- super resolution
- pose estimation
- multiple views
- disparity map
- surface reconstruction
- active learning
- d scene
- stereo images
- pairwise
- background subtraction
- real time
- single image
- vision system
- object detection
- d objects
- dynamic programming
- viewpoint
- object recognition
- computer vision