HRNet: 3D Object Detection Network for Point Cloud with Hierarchical Refinement

Recently, 3D object detection from LiDAR point clouds has advanced rapidly. Although the second
stage can improve the detection performance significantly, prior works concern little about
the essential differences among different stages for the performance enhancement. To address
this, this paper proposes a Hierarchical Refinement Network (HRNet) with two novel strategies.
Firstly, we build the detection head on multi-scale voxel features to optimize the regression
branch progressively with an effective Scale-aware Attentive Propagation (SAP) module. Then,
we propose a Dynamic Sample Selection (DSS) module for the recalculation of the IoU during
each stage to obtain more balanced positive and negative sample selections. Experiments over
benchmark datasets show the effectiveness of our HRNet, particularly for car detection in the
sparse point clouds.