Vertex points are not enough: Monocular 3D object detection via intra- and inter-plane constraints.
Hongdou YaoJun ChenZheng WangXiao WangXiaoyu ChaiYansheng QiuPengfei HanPublished in: Neural Networks (2023)
Keyphrases
- d objects
- pose estimation
- pose parameters
- three dimensional
- viewpoint
- multi view
- geometric invariants
- object recognition
- multiple views
- object detection
- surface points
- viewing direction
- cad models
- point sets
- planar objects
- shape descriptors
- projective geometry
- geometric constraints
- image sequences
- vanishing points
- point distribution
- object contours
- surface patches
- d mesh
- point correspondences
- physically plausible
- pose normalization
- feature points
- range data
- object classes
- trifocal tensor
- computer vision
- calibrated cameras
- camera pose
- topological information
- surface properties
- cad model
- optical flow
- control points
- video sequences
- stereo camera