Context-Aware Data Augmentation for LIDAR 3D Object Detection

TL;DR

This paper introduces CA-aug, a context-aware data augmentation method for LIDAR 3D object detection that improves placement of inserted objects, leading to higher detection accuracy, especially for range-view-based models.

Contribution

The paper proposes CA-aug, a lightweight, context-aware augmentation technique that ensures reasonable object placement, enhancing detection performance over existing methods.

Findings

CA-aug improves mAP by 8% on KITTI dataset.

It outperforms GT-sample and Lidar-aug in detection accuracy.

CA-aug is compatible with other augmentation methods.

Abstract

For 3D object detection, labeling lidar point cloud is difficult, so data augmentation is an important module to make full use of precious annotated data. As a widely used data augmentation method, GT-sample effectively improves detection performance by inserting groundtruths into the lidar frame during training. However, these samples are often placed in unreasonable areas, which misleads model to learn the wrong context information between targets and backgrounds. To address this problem, in this paper, we propose a context-aware data augmentation method (CA-aug) , which ensures the reasonable placement of inserted objects by calculating the "Validspace" of the lidar point cloud. CA-aug is lightweight and compatible with other augmentation methods. Compared with the GT-sample and the similar method in Lidar-aug(SOTA), it brings higher accuracy to the existing detectors. We also present an in-depth study of augmentation methods for the range-view-based(RV-based) models and find that CA-aug can fully exploit the potential of RV-based networks. The experiment on KITTI val split shows that CA-aug can improve the mAP of the test model by 8%.