NeRFool：揭示可泛化神经辐射场面对对抗扰动的脆弱性
NeRFool: Uncovering the Vulnerability of Generalizable Neural Radiance Fields against Adversarial Perturbations
AI总结：
本文首次系统研究可泛化神经辐射场（GNeRF）的对抗鲁棒性，提出NeRFool揭示其脆弱性模式，并开发NeRFool+实现跨目标视图的有效攻击与防御指导，为构建鲁棒GNeRF奠定基础。
摘要：
可泛化神经辐射场（GNeRF）凭借其跨场景泛化能力以及在新场景上即时渲染的可能性，成为新视角合成中最具前景的现实世界解决方案之一。虽然对抗鲁棒性对现实世界应用至关重要，但很少有研究致力于理解其对GNeRF的影响。我们假设，由于GNeRF的实现依赖于来自新场景的源视图条件输入，而这些源视图通常从互联网或第三方提供商获取，因此其现实世界应用可能存在新的安全问题。与此同时，由于GNeRF的3D特性及其独特多样的操作，现有的神经网络对抗鲁棒性理解与解决方案可能并不适用于GNeRF。为此，我们提出了NeRFool，据我们所知，这是首个旨在理解GNeRF对抗鲁棒性的工作。具体而言，NeRFool揭示了GNeRF对抗鲁棒性的脆弱性模式和重要见解。基于从NeRFool中获得的上述见解，我们进一步开发了NeRFool+，它集成了两种能够在大范围目标视图上有效攻击GNeRF的技术，并为防御我们所提出的攻击提供了指导。我们相信，我们的NeRFool/NeRFool+为未来开发鲁棒的现实世界GNeRF解决方案的创新奠定了初步基础。我们的代码可在以下地址获取：https://github.com/GATECH-EIC/NeRFool。
英文摘要：
Generalizable Neural Radiance Fields (GNeRF) are one of the most promising real-world solutions for novel view synthesis, thanks to their cross-scene generalization capability and thus the possibility of instant rendering on new scenes. While adversarial robustness is essential for real-world applications, little study has been devoted to understanding its implication on GNeRF. We hypothesize that because GNeRF is implemented by conditioning on the source views from new scenes, which are often acquired from the Internet or third-party providers, there are potential new security concerns regarding its real-world applications. Meanwhile, existing understanding and solutions for neural networks' adversarial robustness may not be applicable to GNeRF, due to its 3D nature and uniquely diverse operations. To this end, we present NeRFool, which to the best of our knowledge is the first work that sets out to understand the adversarial robustness of GNeRF. Specifically, NeRFool unveils the vulnerability patterns and important insights regarding GNeRF's adversarial robustness. Built upon the above insights gained from NeRFool, we further develop NeRFool+, which integrates two techniques capable of effectively attacking GNeRF across a wide range of target views, and provide guidelines for defending against our proposed attacks. We believe that our NeRFool/NeRFool+ lays the initial foundation for future innovations in developing robust real-world GNeRF solutions. Our codes are available at: https://github.com/GATECH-EIC/NeRFool.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
