arXiv Computer Vision
Scratched Lenses, Shifted Depth: Passive Camera-Side Optical Attacks
Read the original on arXiv Computer Vision →The Flow has not summarised this story yet — read it at arXiv Computer Vision.
Related stories
arXiv Computer Vision
Aug 27
3DGAA: Realistic and Robust 3D Gaussian-based Adversarial Attack for Autonomous Driving
The paper presents 3DGAA, a fabrication-first framework that generates view-consistent, geometry-preserving adversarial wraps for vehicles using 3D Gaussian splatting optimization. It ensures consistency across viewpoints, illumination, and occlusion while limiting changes to vehicle geometry, producing realistic print-only textures that significantly reduce detection confidence and average precision in simulations and physical tests. Ablation and efficiency studies analyze the impact of physical filtering, augmentation, and shape-consistency regularization, and the method demonstrates robustness against common preprocessing defenses and cross-detector transferability.
arXiv AI
Oct 7
Transferable Spatial Temporal Coherence Adversarial Attack on Black-Box Vision Language Models for Autonomous Driving
arXiv:2610.08331v1 Announce Type: cross Abstract: The rapid integration of Vision Language Models (VLMs) into sensitive systems introduces critical safety vulnerabilities that remain unexplored in ex...
arXiv AI
Oct 15
Across Viewpoints and Time: Trajectory-Generalizable Physical Attacks on World Models and Video Understanding Models
arXiv:2610.15343v1 Announce Type: cross Abstract: World models and video understanding models are increasingly deployed in dynamic visual environments, yet their robustness to physical adversarial th...
arXiv Computer Vision
Sep 1
OptiGeo: Efficient Monocular Geometry for Embodied Perception in Optically Challenging Scenes
arXiv:2608.29881v1 Announce Type: new Abstract: Monocular depth estimation has achieved strong open-domain generalization, yet reliable robotic deployment remains difficult in transparent, reflective...
arXiv AI
Jul 7
UNDREAM: Bridging Differentiable Rendering and Photorealistic Simulation for End-to-end Adversarial Attacks
arXiv:2510. 16923v3 Announce Type: replace-cross Abstract: Deep learning models deployed in safety critical applications like autonomous driving use simulations to test their robustness against adversarial attacks in realistic conditions.
arXiv Machine Learning
Sep 10
ComplicitSplat: Downstream Models are Vulnerable to Blackbox Attacks by 3D Gaussian Splat Camouflages
ComplicitSplat is a novel black‑box attack that leverages 3D Gaussian Splatting (3DGS) shading to create viewpoint‑specific camouflage, embedding adversarial content into scene objects that is only visible from certain angles. The method does not require access to model architecture or weights and can successfully fool a range of popular object detectors—including single‑stage, multi‑stage, and transformer‑based models—on both real‑world physical objects and synthetic scenes. This demonstrates that downstream models using 3DGS are vulnerable to adversarial manipulation.