Research map / Adversarial robustness and security / Attacks on text and graphs
Fractured Glass, Failing Cameras: Simulating Physics-Based Adversarial Samples for Autonomous Driving Systems
Manav Prabhakar, Jwalandhar Girnar, Arpan Kusari
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| overlay simulated broken-glass crack on images | YOLOv8 · KITTI (IoU 0.5) | AP, Pedestrian (%): 25.64 -> 17.84 | Hurt |
| overlay simulated crack on images | YOLOv8 · KITTI (IoU 0.5) | AP, Truck (%): 12.39 -> 3.76 | Hurt |
| overlay simulated crack on images | YOLOv8 · KITTI (IoU 0.5) | AP, Car (%): 58.99 -> 57.73 | Hurt |
| overlay simulated crack on images | PVTv2 · BDD100k (IoU 0.5) | AP, Pedestrian (%): 66.47 -> 25.95 | Hurt |
| overlay simulated crack on images | PVTv2 · BDD100k (IoU 0.5) | AP, Car (%): 80.37 -> 56.78 | Hurt |
| overlay simulated crack on images | Faster R-CNN · MS-COCO (IoU 0.5) | AP, Food (%): 35.34 -> 30.65 | Hurt |
| overlay real online crack on images | YOLOv8 · KITTI (IoU 0.5) | AP, Pedestrian (%): 25.64 -> 69.72 | Improved |
| overlay real online crack on images | YOLOv8 · KITTI (IoU 0.5) | AP, Truck (%): 12.39 -> 3.59 | Hurt |
| overlay simulated crack on images | YOLOv8 · KITTI (IoU 0.75) | AP, Pedestrian (%): 6.83 -> 6.02 | No clear effect |
| use simulated cracks in place of real online cracks as adversarial filter (ablation) | detectors · KITTI/BDD100k/MS-COCO | AP drop | Mixed |
Closest papers by meaning
- PhysGAN: Generating Physical-World-Resilient Adversarial Examples for Autonomous Driving (CVPR 2020 · similarity 0.85)
- Understanding the Robustness of 3D Object Detection With Bird's-Eye-View Representations in Autonomous Driving (CVPR 2023 · similarity 0.85)
- 3D-VField: Adversarial Augmentation of Point Clouds for Domain Generalization in 3D Object Detection (CVPR 2022 · similarity 0.85)
- SplatAD: Real-Time Lidar and Camera Rendering with 3D Gaussian Splatting for Autonomous Driving (CVPR 2025 · similarity 0.84)
- RLGF: Reinforcement Learning with Geometric Feedback for Autonomous Driving Video Generation (NeurIPS 2025 · similarity 0.83)
- Towards Visual Discrimination and Reasoning of Real-World Physical Dynamics: Physics-Grounded Anomaly Detection (CVPR 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
