DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection.
Liang PengXiaopei WuZheng YangHaifeng LiuDeng CaiPublished in: CoRR (2022)
Keyphrases
- d objects
- pose estimation
- object recognition
- viewpoint
- multiple views
- depth cues
- multi view
- three dimensional
- shape descriptors
- range data
- cad models
- image sequences
- depth map
- depth images
- pose normalization
- geometric models
- partial matching
- human body
- line drawings
- object classes
- hough transform
- object features
- image contours
- computer vision
- object views
- machine learning