Real-Time Moving Object Detection for 3-D LiDAR Using Occlusion Accumulation in Range Image

Abstract

Deep learning methods have been applied to detect moving objects from 3-D LiDAR data, but they require extra computing devices, such as graphics processing units (GPUs), and often struggle to operate at real-time LiDAR frame rates. In addition, retraining is necessary for new environments, which demands significant resources. Nonlearning methods that operate online without prior maps and rely solely on a central processing unit (CPU) often fail to achieve real-time performance. Furthermore, they are limited in applicability due to their dependence on specific pose estimation algorithms. To address these issues, we propose a novel occlusion accumulation framework in the range image domain for real-time 3-D LiDAR moving object detection on a CPU. By incorporating compensation strategies that account for LiDAR artifacts and measurement sparsity, the proposed method reduces false positives (FPs) and improves detection accuracy. Our approach is also flexible, integrating seamlessly with different pose estimation algorithms without performance degradation. Extensive experiments on KITTI, Apollo, and CARLA datasets demonstrate that the proposed method achieves competitive results compared to state-of-the-art (SOTA) learning-based methods on KITTI and outperforms them in cross-validation on Apollo and CARLA. Moreover, the proposed method operates four times faster than SOTA nonlearning-based methods while offering higher detection accuracy, making it highly suitable for real-world applications in autonomous driving and robotic navigation. We provide all the source code and datasets used in this article to the public at: <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/JunhaAgu/Mapless_Moving</uri>.