Patch Refinement -- Localized 3D Object Detection

We introduce Patch Refinement a two-stage model for accurate 3D object detection and localization from point cloud data. Patch Refinement is composed of two independently trained Voxelnet-based networks, a Region Proposal Network (RPN) and a Local Refinement Network (LRN). We decompose the detection task into a preliminary Bird's Eye View (BEV) detection step and a local 3D detection step. Based on the proposed BEV locations by the RPN, we extract small point cloud subsets ("patches"), which are then processed by the LRN, which is less limited by memory constraints due to the small area of each patch. Therefore, we can apply encoding with a higher voxel resolution locally. The independence of the LRN enables the use of additional augmentation techniques and allows for an efficient, regression focused training as it uses only a small fraction of each scene. Evaluated on the KITTI 3D object detection benchmark, our submission from January 28, 2019, outperformed all previous entries on all three difficulties of the class car, using only 50 % of the available training data and only LiDAR information.

PDF Abstract

Datasets


Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Object Detection KITTI Cars Easy Patches AP 87.87 # 1
Birds Eye View Object Detection KITTI Cars Easy Patches AP 89.78 # 4
Object Detection KITTI Cars Hard Patches AP 68.91 # 1
Birds Eye View Object Detection KITTI Cars Hard Patches AP 79.22 # 6
Object Detection KITTI Cars Moderate Patches AP 77.16 # 1
Birds Eye View Object Detection KITTI Cars Moderate Patches AP 86.55% # 5

Methods