Sparse Semantic Map-Based Monocular Localization in Traffic Scenes Using Learned 2D-3D Point-Line Correspondences.
Xingyu ChenJianru XueShanmin PangPublished in: CoRR (2022)
Keyphrases
- line correspondences
- pose estimation
- traffic scenes
- road vehicles
- image sequences
- human body
- pedestrian detection
- camera pose
- computer vision
- d objects
- structure from motion
- moving vehicles
- position and orientation
- multiple views
- feature points
- video sequences
- point correspondences
- ground plane
- motion analysis
- trifocal tensor
- multiple cameras
- focal length
- image features