DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection.
Liang PengXiaopei WuZheng YangHaifeng LiuDeng CaiPublished in: ECCV (1) (2022)
Keyphrases
- d objects
- pose estimation
- viewpoint
- multi view
- object recognition
- shape descriptors
- cad models
- three dimensional
- object classes
- depth images
- depth map
- object detection
- multiple views
- depth cues
- depth information
- object representation
- image contours
- d mesh
- range data
- detection method
- image sequences
- line drawings
- machine learning
- geometric models
- optical flow
- object views
- human body
- object features
- pose normalization