SVMAC: Unsupervised 3D Human Pose Estimation from a Single Image with Single-view-multi-angle Consistency.
Yicheng DengCheng SunJiahui ZhuYongqi SunPublished in: 3DV (2021)
Keyphrases
- single image
- single view
- multi view
- vanishing points
- perspective projection
- shape recovery
- multiple views
- multiple images
- image pairs
- light source
- surface reconstruction
- d scene
- shape from shading
- super resolution
- semi supervised
- three dimensional
- calibrated cameras
- indoor scenes
- view invariant
- spherical harmonics
- body parts
- range data
- action recognition
- active learning
- pairwise
- machine learning