A Dataset-Dispersion Perspective on Reconstruction Versus Recognition in Single-View 3D Reconstruction Networks.
Yefan ZhouYiru ShenYujun YanChen FengYaoqing YangPublished in: 3DV (2021)
Keyphrases
- single view
- multi view
- single view reconstruction
- three dimensional
- visual hull
- perspective projection
- multiple views
- motion stereo
- polyhedral objects
- co training
- vanishing points
- object recognition
- urban scenes
- surface reconstruction
- feature extraction
- reconstruction process
- viewpoint
- d objects
- high resolution
- camera calibration
- repetitive structures
- shape recovery
- stereo reconstruction
- non rigid structure from motion
- training set
- structure from motion
- piecewise planar
- view invariant
- training data
- human actions
- activity recognition