MonoJSG: Joint Semantic and Geometric Cost Volume for Monocular 3D Object Detection.
Qing LianPeiliang LiXiaozhi ChenPublished in: CoRR (2022)
Keyphrases
- d objects
- pose estimation
- geometric models
- viewpoint
- multi view
- multiple views
- object recognition
- shape descriptors
- partial matching
- object parts
- range data
- three dimensional
- d mesh
- cad model
- cad models
- integral imaging
- topological information
- object classes
- image sequences
- object detection
- pose normalization
- physically plausible
- spin images
- recognition and pose estimation
- three dimensional objects
- line drawings
- image contours
- surface points
- real world objects
- object views
- complex background
- stereo camera
- human body