Cai Mumuxin, Wang Xupeng, Sohel Ferdous, Lei Hang
School of Information and Software Engineering, The University of Electronic Science and Technology of China, Chengdu 610054, China.
Laboratory Of Intelligent Collaborative Computing, The University of Electronic Science and Technology ofChina, Chengdu 610054, China.
Sensors (Basel). 2024 Aug 22;24(16):5440. doi: 10.3390/s24165440.
LiDAR sensors have been shown to generate data with various common corruptions, which seriously affect their applications in 3D vision tasks, particularly object detection. At the same time, it has been demonstrated that traditional defense strategies, including adversarial training, are prone to suffering from gradient confusion during training. Moreover, they can only improve their robustness against specific types of data corruption. In this work, we propose LiDARPure, which leverages the powerful generation ability of diffusion models to purify corruption in the LiDAR scene data. By dividing the entire scene into voxels to facilitate the processes of diffusion and reverse diffusion, LiDARPure overcomes challenges induced from adversarial training, such as sparse point clouds in large-scale LiDAR data and gradient confusion. In addition, we utilize the latent geometric features of a scene as a condition to assist the generation of diffusion models. Detailed experiments show that LiDARPure can effectively purify 19 common types of LiDAR data corruption. Further evaluation results demonstrate that it can improve the average precision of 3D object detectors to an extent of 20% in the face of data corruption, much higher than existing defence strategies.
激光雷达传感器已被证明会生成带有各种常见损坏的数据，这严重影响了它们在三维视觉任务中的应用，尤其是目标检测。同时，已经证明传统的防御策略，包括对抗训练，在训练过程中容易遭受梯度混淆。此外，它们只能提高对特定类型数据损坏的鲁棒性。在这项工作中，我们提出了LiDARPure，它利用扩散模型强大的生成能力来净化激光雷达场景数据中的损坏。通过将整个场景划分为体素以促进扩散和反向扩散过程，LiDARPure克服了对抗训练带来的挑战，如大规模激光雷达数据中的稀疏点云和梯度混淆。此外，我们利用场景的潜在几何特征作为条件来辅助扩散模型的生成。详细实验表明，LiDARPure可以有效地净化19种常见类型的激光雷达数据损坏。进一步的评估结果表明，面对数据损坏时，它可以将三维目标检测器的平均精度提高20%，远高于现有的防御策略。
