The Devil is in the Task: Exploiting Reciprocal Appearance-Localization Features for Monocular 3D Object Detection.
Zhikang ZouXiaoqing YeLiang DuXianhui ChengXiao TanLi ZhangJianfeng FengXiangyang XueErrui DingPublished in: CoRR (2021)
Keyphrases
- d objects
- pose estimation
- surface features
- object features
- object representation
- multi view
- object recognition
- object model
- multiple views
- cad models
- viewpoint
- object parts
- range data
- pose normalization
- occluded objects
- machine learning
- multi views
- line drawings
- object detection
- image features
- feature vectors
- three dimensional
- salient features
- d mesh
- static images
- structural information
- spatial information
- face detection
- feature points
- surface properties