Object DGCNN: 3D Object Detection using Dynamic Graphs
This addresses the problem of simplifying complex pipelines in 3D object detection for autonomous driving applications, though it appears incremental as it builds on existing graph-based frameworks.
The paper tackles 3D object detection from point clouds by proposing a method that models detection as message passing on a dynamic graph, eliminating the need for post-processing like non-maximum suppression, and achieves state-of-the-art performance on autonomous driving benchmarks.
3D object detection often involves complicated training and testing pipelines, which require substantial domain knowledge about individual datasets. Inspired by recent non-maximum suppression-free 2D object detection models, we propose a 3D object detection architecture on point clouds. Our method models 3D object detection as message passing on a dynamic graph, generalizing the DGCNN framework to predict a set of objects. In our construction, we remove the necessity of post-processing via object confidence aggregation or non-maximum suppression. To facilitate object detection from sparse point clouds, we also propose a set-to-set distillation approach customized to 3D detection. This approach aligns the outputs of the teacher model and the student model in a permutation-invariant fashion, significantly simplifying knowledge distillation for the 3D detection task. Our method achieves state-of-the-art performance on autonomous driving benchmarks. We also provide abundant analysis of the detection model and distillation framework.