CAGroup3D: Class-Aware Grouping for 3D Object Detection on Point Clouds
This addresses 3D object detection for robotics and autonomous systems, showing incremental but strong gains over existing methods.
The paper tackles 3D object detection on point clouds by introducing a two-stage fully sparse convolutional framework that uses class-aware grouping and sparse RoI pooling, achieving state-of-the-art performance with gains of 3.6% mAP@0.25 on ScanNet V2 and 2.6% on SUN RGB-D.
We present a novel two-stage fully sparse convolutional 3D object detection framework, named CAGroup3D. Our proposed method first generates some high-quality 3D proposals by leveraging the class-aware local group strategy on the object surface voxels with the same semantic predictions, which considers semantic consistency and diverse locality abandoned in previous bottom-up approaches. Then, to recover the features of missed voxels due to incorrect voxel-wise segmentation, we build a fully sparse convolutional RoI pooling module to directly aggregate fine-grained spatial information from backbone for further proposal refinement. It is memory-and-computation efficient and can better encode the geometry-specific features of each 3D proposal. Our model achieves state-of-the-art 3D detection performance with remarkable gains of +\textit{3.6\%} on ScanNet V2 and +\textit{2.6}\% on SUN RGB-D in term of mAP@0.25. Code will be available at https://github.com/Haiyang-W/CAGroup3D.