Neural Radiance Fields (NeRFs) have recently emerged as a powerful tool for 3D scene representation and rendering. These data-driven models can learn to synthesize high-quality images from sparse 2D observations, enabling realistic and interactive scene reconstructions. However, the growing usage of NeRFs in critical applications such as augmented reality, robotics, and virtual environments could be threatened by adversarial attacks. In this paper we present how generalizable NeRFs can be attacked by both low-intensity adversarial attacks and adversarial patches, where the later could be robust enough to be used in real world applications. We also demonstrate targeted attacks, where a specific, predefined output scene is generated by these attack with success.

翻译：神经辐射场（NeRF）近期已成为三维场景表示与渲染的有力工具。这类数据驱动的模型能够从稀疏的二维观测中学习合成高质量图像，从而实现逼真且可交互的场景重建。然而，随着NeRF在增强现实、机器人技术及虚拟环境等关键应用中的日益普及，其可能面临对抗攻击的威胁。本文展示了如何通过低强度对抗攻击和对抗性补丁两种方式攻击可泛化NeRF，其中后者具备足够鲁棒性，可应用于现实场景。我们同时验证了定向攻击的有效性——此类攻击能够成功生成预定义的特定输出场景。