Geometry-Biased Transformer for Robust Multi-View 3D Human Pose Reconstruction.
Olivier MolinerSangxia HuangKalle ÅströmPublished in: FG (2024)
Keyphrases
- multi view
- multi view stereo
- three dimensional
- scene reconstruction
- single view
- multiple views
- depth map
- camera calibration
- visual hull
- multiple cameras
- d objects
- dynamic scenes
- semi supervised
- view dependent
- multi view clustering
- view synthesis
- bundle adjustment
- multi view images
- surface model
- shape from silhouettes
- camera parameters
- multi view face detection
- multi view reconstruction
- multi view learning
- multiple viewpoints
- geometric constraints
- high quality
- structure from motion
- light field
- surface reconstruction
- motion capture
- d scene
- point cloud
- image reconstruction
- range images
- scene structure
- high resolution
- aam fitting
- reconstruction method
- object recognition