BKinD-3D: Self-Supervised 3D Keypoint Discovery from Multi-View Videos.
Jennifer J. SunPierre KarashchukAmil DravidSerim RyouSonia FereidooniJohn TuthillAggelos K. KatsaggelosBingni W. BruntonGeorgia GkioxariAnn KennedyYisong YuePietro PeronaPublished in: CoRR (2022)
Keyphrases
- multi view
- multi view video sequences
- dynamic scenes
- single view
- multiple views
- view invariant
- depth map
- three dimensional
- keypoints
- camera calibration
- multiple cameras
- d objects
- multi view clustering
- multi view reconstruction
- video sequences
- multi view stereo
- semi supervised
- multiple viewpoints
- range images
- video surveillance
- image patches
- video content
- co training
- multi view face detection
- image features
- video frames