Synthesizing 3D Shapes via Modeling Multi-view Depth Maps and Silhouettes with Deep Generative Networks.
Amir Arsalan SoltaniHaibin HuangJiajun WuTejas D. KulkarniJoshua B. TenenbaumPublished in: CVPR (2017)
Keyphrases
- multi view
- depth map
- single view
- depth information
- d objects
- d scene
- view synthesis
- multiple views
- stereo matching
- visual hull
- camera calibration
- multi view stereo
- three dimensional
- dynamic scenes
- free viewpoint
- semi supervised
- depth estimation
- photometric stereo
- real scenes
- scene reconstruction
- multi view images
- multiple cameras
- bundle adjustment
- view dependent
- learning algorithm
- surface reconstruction
- depth data
- shape model
- range images
- virtual view synthesis
- shape from silhouettes
- stereo pair
- camera parameters
- background subtraction
- motion estimation
- depth video