On the Importance of Stereo for Accurate Depth Estimation: An Efficient Semi-Supervised Deep Neural Network Approach.
Nikolai SmolyanskiyAlexey KamenevStan BirchfieldPublished in: CoRR (2018)
Keyphrases
- depth estimation
- semi supervised
- depth map
- stereo vision
- depth estimates
- binocular stereo
- stereo matching
- multi view
- depth information
- motion parallax
- dynamic scenes
- stereo pair
- scene understanding
- disparity map
- depth perception
- dense depth maps
- high quality
- range sensors
- super resolution
- real scenes
- stereo images
- active learning
- blur estimation
- d scene
- object detection
- input image
- feature matching
- object recognition
- depth from defocus
- object boundaries
- real time
- multiple views
- space time
- high resolution
- three dimensional
- image processing