Research map / Adversarial robustness and security / Attacks on text and graphs
Cheating Stereo Matching in Full-Scale: Physical Adversarial Attack Against Binocular Depth Estimation in Autonomous Driving
Kangqiao Zhao, Shuo Huai, Xurui Song, Jun Luo
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 |
|---|---|---|---|
| add stereo-aligned 3D rendering (SAR) (ablation) | merging attack · CARLA (PSMNet) | E_cover: 0.019 -> 0.573 (merge only vs full) | Improved |
| add merging loss (ablation) | merging attack with SAR · CARLA (PSMNet) | E_blend: 0.411 -> 0.058 | Improved |
| remove non-printability loss (ablation) | merging attack · CARLA (PSMNet) | E_blend: 0.058 -> 0.042 | No clear effect |
| replace 2D patch with full-surface 3D texture | PASM/Adv-DM · CARLA, 5 stereo models | E_cover: 0.154 -> 0.553 (PASM vs ours, PSMNet) | Improved |
| texture trained on PSMNet transferred to RAFT-Stereo | merging attack · CARLA | E_blend: 0.087 (GA-Net) vs 0.128 (RAFT-Stereo) | Mixed |
| full Apollo stack with adversarial vehicle | merging attack · Apollo collision rate | E_sys: 0 -> 0.74 (PSMNet) | Improved |
Closest papers by meaning
- DepthVanish: Optimizing Adversarial Interval Structures for Stereo-Depth-Invisible Patches (NeurIPS 2025 · similarity 0.92)
- Physical 3D Adversarial Attacks against Monocular Depth Estimation in Autonomous Driving (CVPR 2024 · similarity 0.92)
- Physical Attack on Monocular Depth Estimation with Optimal Adversarial Patches (ECCV 2022 · similarity 0.91)
- Beware of Road Markings: A New Adversarial Patch Attack to Monocular Depth Estimation (NeurIPS 2024 · similarity 0.90)
- Stereoscopic Universal Perturbations Across Different Architectures and Datasets (CVPR 2022 · similarity 0.88)
- Stereo Risk: A Continuous Modeling Approach to Stereo Matching (ICML 2024 · similarity 0.85)
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.
