False Positive Removal for 3D Vehicle Detection with Penetrated Point Classifier
This work addresses false positives in 3D vehicle detection for autonomous driving systems, representing an incremental improvement by enhancing an existing state-of-the-art method.
The paper tackled the problem of false positive boxes in 3D vehicle detection from LiDAR point clouds, which occur at high recall positions due to few points on objects, and introduced a Penetrated Point Classifier that improved precision by 15.46 and 14.63 percentage points on moderate and hard difficulty levels in the KITTI dataset.
Recently, researchers have been leveraging LiDAR point cloud for higher accuracy in 3D vehicle detection. Most state-of-the-art methods are deep learning based, but are easily affected by the number of points generated on the object. This vulnerability leads to numerous false positive boxes at high recall positions, where objects are occasionally predicted with few points. To address the issue, we introduce Penetrated Point Classifier (PPC) based on the underlying property of LiDAR that points cannot be generated behind vehicles. It determines whether a point exists behind the vehicle of the predicted box, and if does, the box is distinguished as false positive. Our straightforward yet unprecedented approach is evaluated on KITTI dataset and achieved performance improvement of PointRCNN, one of the state-of-the-art methods. The experiment results show that precision at the highest recall position is dramatically increased by 15.46 percentage points and 14.63 percentage points on the moderate and hard difficulty of car class, respectively.