Depth Estimation From a Light Field Image Pair With a Generative Model.
Tao YanFan ZhangYiming MaoHongbin YuXiaohua QianRynson W. H. LauPublished in: IEEE Access (2019)
Keyphrases
- generative model
- light field
- depth estimation
- image pairs
- super resolution
- single image
- disparity map
- feature matching
- multi view
- depth map
- stereo images
- stereo pair
- epipolar geometry
- probabilistic model
- high resolution
- low resolution
- stereo vision
- dynamic scenes
- image formation
- stereo matching
- image matching
- image registration
- input image
- semi supervised
- real scenes
- depth information
- multiple images
- em algorithm
- high quality
- image restoration
- view synthesis
- d scene
- multiple views
- motion blur
- three dimensional
- video frames
- motion estimation
- ground truth
- training data
- belief propagation
- light source
- expectation maximization
- pattern recognition
- computer vision