GitHub repository
A 3D object detection method that exploits visibility information from LiDAR point clouds to improve accuracy.
What is GitHub repository?
WYSIWYG is a 3D object detection method that leverages visibility information from LiDAR point clouds to improve detection accuracy. It addresses the problem of detecting objects in 3D environments where occlusion and partial visibility are common, such as in autonomous driving scenarios. The approach integrates raycasting to model visibility and feeds this into a deep learning detector for enhanced performance.
Researchers and engineers working on 3D computer vision, particularly in autonomous driving, robotics, or any field requiring accurate object detection from LiDAR data.
Developers choose WYSIWYG for its novel use of visibility cues, which leads to state-of-the-art detection accuracy on benchmarks like NuScenes. It provides an open-source implementation built on the established SECOND framework, making it accessible for experimentation and integration.
Overview
WYSIWYG: Exploiting Visibility for 3D Object Detection, CVPR 2020 (Oral)
Use Cases
Best For
- Improving 3D object detection accuracy in occluded urban environments
Related Projects
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
