Enhancing Self-Supervised Monocular Depth Estimation via Piece-Wise Pose Estimation and Geometric Constraints.
Pranjay ShyamAlexandre OkonHyunjin YooPublished in: WACV (Workshops) (2024)
Keyphrases
- pose estimation
- geometric constraints
- depth estimation
- feature matching
- stereo pair
- epipolar geometry
- feature points
- stereo vision
- depth map
- camera calibration
- stereo matching
- computer vision
- position and orientation
- d objects
- depth information
- dynamic scenes
- human body
- real scenes
- human pose
- multiple images
- depth images
- super resolution
- visual hull
- multi view
- image sequences
- focal length
- multiple cameras
- body parts
- ground plane
- video sequences
- disparity map
- machine learning
- high resolution
- object boundaries