AutoAlign: Pixel-Instance Feature Aggregation for Multi-Modal 3D Object Detection.
Zehui ChenZhenyu LiShiquan ZhangLiangji FangQinhong JiangFeng ZhaoBolei ZhouHang ZhaoPublished in: IJCAI (2022)
Keyphrases
- multi modal
- d objects
- object features
- object recognition
- multi view
- pose estimation
- viewpoint
- cad models
- object detection
- multiple views
- multi modality
- range data
- point distribution
- three dimensional
- d mesh
- cross modal
- physically plausible
- shape descriptors
- feature vectors
- image contours
- image features
- topological information
- high dimensional
- multi views
- integral imaging
- single modality
- object classes
- image annotation
- face detection