ComplicitSplat: Downstream Models are Vulnerable to Blackbox Attacks by 3D Gaussian Splat Camouflages

TL;DR

This paper presents ComplicitSplat, a novel black-box attack exploiting 3D Gaussian Splatting to embed viewpoint-specific adversarial camouflage, successfully fooling various object detectors and revealing safety vulnerabilities in critical systems.

Contribution

Introduces the first black-box attack leveraging 3D Gaussian Splatting to create viewpoint-dependent adversarial camouflage for object detectors.

Findings

Successfully attacks multiple detector types

Effective on real-world and synthetic scenes

Reveals safety risks in autonomous systems

Abstract

As 3D Gaussian Splatting (3DGS) gains rapid adoption in safety-critical tasks for efficient novel-view synthesis from static images, how might an adversary tamper images to cause harm? We introduce ComplicitSplat, the first attack that exploits standard 3DGS shading methods to create viewpoint-specific camouflage - colors and textures that change with viewing angle - to embed adversarial content in scene objects that are visible only from specific viewpoints and without requiring access to model architecture or weights. Our extensive experiments show that ComplicitSplat generalizes to successfully attack a variety of popular detector - both single-stage, multi-stage, and transformer-based models on both real-world capture of physical objects and synthetic scenes. To our knowledge, this is the first black-box attack on downstream object detectors using 3DGS, exposing a novel safety risk for applications like autonomous navigation and other mission-critical robotic systems.