Point Cloud Completion of Foot Shape from a Single Depth Map for Fit Matching Using Deep Learning View Synthesis.
John S. ZelekNolan LunscherPublished in: ICCV Workshops (2017)
Keyphrases
- view synthesis
- depth map
- point cloud
- photorealistic
- multi view
- d scene
- depth information
- stereo camera
- high quality
- three dimensional
- stereo matching
- free viewpoint
- structure from motion
- depth estimation
- object boundaries
- image sequences
- stereo correspondence
- matching algorithm
- depth image based rendering
- virtual view
- surface reconstruction
- point sets
- multiple cameras
- disparity map
- stereo pair
- super resolution
- viewpoint
- dynamic scenes
- optical flow
- input image
- machine learning
- depth images
- single image
- camera calibration