3DRef: 3D Dataset and Benchmark for Reflection Detection in RGB and Lidar Data
This addresses a persistent problem for robotics and autonomous systems by providing a comprehensive dataset, though it is incremental as it extends existing 2D reflection detection to 3D.
The paper tackles the challenge of detecting reflective surfaces in 3D environments by introducing the first large-scale 3D reflection detection dataset with over 50,000 aligned samples of Lidar and RGB data, and it benchmarks existing segmentation methods to advance research in this area.
Reflective surfaces present a persistent challenge for reliable 3D mapping and perception in robotics and autonomous systems. However, existing reflection datasets and benchmarks remain limited to sparse 2D data. This paper introduces the first large-scale 3D reflection detection dataset containing more than 50,000 aligned samples of multi-return Lidar, RGB images, and 2D/3D semantic labels across diverse indoor environments with various reflections. Textured 3D ground truth meshes enable automatic point cloud labeling to provide precise ground truth annotations. Detailed benchmarks evaluate three Lidar point cloud segmentation methods, as well as current state-of-the-art image segmentation networks for glass and mirror detection. The proposed dataset advances reflection detection by providing a comprehensive testbed with precise global alignment, multi-modal data, and diverse reflective objects and materials. It will drive future research towards reliable reflection detection. The dataset is publicly available at http://3dref.github.io