-MonoDETR: Supervised Shape&Scale-perceptive Deformable Transformer for Monocular 3D Object Detection.
Xuan HeKailun YangJunwei ZhengJin YuanLuis Miguel BergasaHui ZhangZhiyong LiPublished in: CoRR (2023)
Keyphrases
- d objects
- shape descriptors
- pose estimation
- cad models
- image contours
- d mesh
- object recognition
- curved objects
- geometric models
- topological information
- viewpoint
- three dimensional
- recognition and pose estimation
- surface properties
- multi view
- range data
- object features
- shape matching
- object contours
- image sequences
- object representation
- surface patches
- multiple views
- shape model
- object classes
- partial matching
- surface points
- line drawings
- articulated objects
- pose normalization
- shape representation
- cad model
- object matching
- spin images
- object detection
- shape analysis
- computer vision
- physically plausible
- point distribution
- semi supervised
- three dimensional objects
- stereo camera
- surface reconstruction