UMVpose++: Unsupervised Multi-View Multi-Person 3D Pose Estimation Using Ground Point Matching.
Diógenes Wallis de França SilvaJoão Paulo Silva do Monte LimaDiego Gabriel Francis ThomasHideaki UchiyamaVeronica TeichriebPublished in: VISIGRAPP (4: VISAPP) (2023)
Keyphrases
- multi view
- point matching
- semi supervised
- multiple views
- fundamental matrix
- point correspondences
- matching algorithm
- feature points
- shape recognition
- single view
- point sets
- d objects
- pose estimation
- graph matching
- control points
- depth map
- three dimensional
- camera calibration
- affine invariant
- supervised learning
- range images
- viewpoint
- pairwise
- surface reconstruction
- epipolar constraint
- computer vision
- d scene
- articulated objects