Self-supervised Monocular Trained Depth Estimation using Self-attention and Discrete Disparity Volume.
Adrian JohnstonGustavo CarneiroPublished in: CoRR (2020)
Keyphrases
- depth estimation
- monocular images
- stereo matching
- disparity map
- stereo pair
- depth cues
- stereo vision
- depth map
- stereo algorithms
- stereo images
- stereo camera
- binocular stereo
- stereo correspondence
- stereo algorithm
- depth information
- depth estimates
- image sequences
- scene flow
- ground truth
- image pairs
- disparity estimation
- real scenes
- pose estimation
- dynamic programming
- dynamic scenes
- multi view
- belief propagation
- surface reconstruction
- feature matching
- scene understanding
- d scene
- super resolution
- specular reflection
- position and orientation
- object boundaries
- point cloud
- computer vision
- vision system
- ego motion
- input image
- epipolar geometry
- post processing
- real time
- three dimensional