Robust environment perception is critical for autonomous cars, and adversarial defenses are the most effective and widely studied ways to improve the robustness of environment perception. However, all of previous defense methods decrease the natural accuracy, and the nature of the DNNs itself has been overlooked. To this end, in this paper, we propose a novel adversarial defense for 3D point cloud classifier that makes full use of the nature of the DNNs. Due to the disorder of point cloud, all point cloud classifiers have the property of permutation invariant to the input point cloud. Based on this nature, we design invariant transformations defense (IT-Defense). We show that, even after accounting for obfuscated gradients, our IT-Defense is a resilient defense against state-of-the-art (SOTA) 3D attacks. Moreover, IT-Defense do not hurt clean accuracy compared to previous SOTA 3D defenses. Our code is available at: {\footnotesize{\url{https://github.com/cuge1995/IT-Defense}}}.
翻译:强力环境感知对于自主汽车至关重要,而对抗性防守是提高环境感知稳健性的最有效和广泛研究的方法。 但是,所有先前的防守方法都降低了自然精确度,DNNs本身也被忽视了。 为此,我们在本文件中提议为3D点云分分类器提供新的对抗性防守,充分利用DNNs的性质。由于点云的混乱,所有点云分级器都具有输入点云的变异特性。基于这种性质,我们设计了不易变换的防(IT-Defence ) 。我们显示,即使在计算了模糊的梯度之后,我们的IT-Defence 也是抵抗最先进的(SOTA) 3D 攻击的有弹性的防御。 此外,与以前的SOTA 3D防御相比, IT- Defreat并不伤害干净的准确性。我们的代码可以在以下查阅: whootootardisucult {http://github.com/coldion1995/IT-defilen 。