Rethinking Training Objective For Self-Supervised Monocular Depth Estimation: Semantic Cues To Rescue.
Keyao LiGe LiThomas H. LiPublished in: ICIP (2021)
Keyphrases
- depth estimation
- depth cues
- depth perception
- monocular images
- stereo vision
- depth map
- depth estimates
- depth information
- stereo matching
- real scenes
- scene understanding
- image sequences
- model based pose estimation
- stereo pair
- dynamic scenes
- training set
- super resolution
- disparity map
- multi view
- object boundaries
- d scene
- multiple views
- high resolution
- high quality
- computer vision