ImGeoNet: Image-induced Geometry-aware Voxel Representation for Multi-view 3D Object Detection.
Tao TuShun-Po ChuangYu-Lun LiuCheng SunKe ZhangDonna RoyCheng-Hao KuoMin SunPublished in: ICCV (2023)
Keyphrases
- multi view
- d objects
- multi views
- three dimensional
- multiple views
- single view
- image data
- geometric models
- depth map
- input image
- topological information
- photometric stereo
- camera calibration
- single image
- object recognition
- shape descriptors
- test images
- range data
- viewpoint
- image features
- scene geometry
- pose estimation
- object detection
- semi supervised
- camera parameters
- light field
- pose parameters
- geometric constraints
- high resolution
- viewing direction
- visual hull
- view synthesis
- d mesh
- bundle adjustment
- surface patches
- image matching
- feature points
- image sequences