CAPE: Camera View Position Embedding for Multi-View 3D Object Detection.
Kaixin XiongShi GongXiaoqing YeXiao TanJi WanErrui DingJingdong WangXiang BaiPublished in: CoRR (2023)
Keyphrases
- multi view
- d objects
- camera views
- multiple views
- multiple cameras
- dynamic scenes
- camera calibration
- single view
- pose estimation
- object recognition
- depth map
- viewpoint
- object detection
- three dimensional
- range data
- multi camera
- camera parameters
- ground plane
- multi views
- object classes
- surface reconstruction
- point correspondences
- visual hull
- d scene
- d mesh
- computer vision
- texture mapping
- machine learning