CMDA: Cross-Modal and Domain Adversarial Adaptation for LiDAR-Based 3D Object Detection.
Gyusam ChangWonseok RohSujin JangDongwook LeeDaehyun JiGyeongrok OhJinsun ParkJinkyu KimSangpil KimPublished in: CoRR (2024)
Keyphrases
- d objects
- cross modal
- viewpoint
- object recognition
- multi view
- multi modal
- pose estimation
- three dimensional
- multiple views
- visual recognition
- object classes
- object detection
- multimedia databases
- multimedia retrieval
- cad models
- contextual information
- image data
- human body
- high resolution
- point cloud
- human motion
- multimedia