Research map / Adversarial robustness and security / Attacks on text and graphs
MOBA: A Material-Oriented Backdoor Attack Against LiDAR-Based 3D Object Detection Systems
Saket Sanjeev Chaturvedi, Gaurav Bagwe, Lan Emily Zhang, Pan He, Xiaoyong Yuan
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 angle-robust (angle-independent Oren-Nayar) intensity simulation (ablation) | MOBA · Physical trigger dataset, resizing attack | ASR (%): 53.06 -> 90.23 (PointPillars) | Improved |
| add distance-robust trigger scaling (ablation) | MOBA · Physical trigger dataset, resizing attack | ASR (%): 79.59 -> 94.00 (SECOND) | Improved |
| add angle-robust intensity simulation (ablation) | MOBA · Physical trigger dataset, resizing attack, VoxelNet | ASR (%): 84.35 -> 93.87 | Improved |
| add distance-robust scaling (ablation) | MOBA · Physical trigger dataset, resizing attack, MVX-Net | ASR (%): 79.59 -> 95.91 | Improved |
| use TiO2 as trigger material instead of copper (ablation) | MOBA · Trigger material, MVX-Net | ASR (%): 78.43 -> 95.91 | Improved |
| use TiO2 as trigger material instead of aluminum (ablation) | MOBA · Trigger material, MVX-Net | ASR (%): 37.25 -> 95.91 | Improved |
| use reflective TiO2 trigger instead of visually similar white paper (ablation) | MOBA · Trigger material, MVX-Net | ASR (%): 46.05 -> 95.91 | Improved |
| use BRDF-based intensity instead of fixed intensity 0.5 (ablation) | MOBA · Intensity simulation, MVX-Net | ASR (%): 79.59 -> 95.91 | Improved |
| use BRDF-based intensity instead of random intensity (ablation) | MOBA · Intensity simulation, MVX-Net | ASR (%): 91.83 -> 95.91 | Improved |
| include BRDF-based intensity instead of no intensity (ablation) | MOBA · Intensity simulation, MVX-Net | ASR (%): 81.63 -> 95.91 | Improved |
| switch attack objective from resizing to disappearance (ablation) | MOBA · VoxelNet | ASR (%): 93.87 -> 95.33 | Improved |
| use MOBA instead of BadFusion on fusion model | MVX-Net · Physical trigger dataset, resizing attack | ASR (%): 63.26 -> 95.91 | Improved |
Closest papers by meaning
- MOS: Model Synergy for Test-Time Adaptation on LiDAR-Based 3D Object Detection (ICLR 2025 · similarity 0.85)
- An Empirical Study of the Generalization Ability of Lidar 3D Object Detectors to Unseen Domains (CVPR 2024 · similarity 0.83)
- Neural Reconstruction of LiDAR Point Clouds under Jamming Attacks via Full-Waveform Representation and Simultaneous Laser Sensing (NeurIPS 2026 · similarity 0.83)
- SimDistill: Simulated Multi-Modal Distillation for BEV 3D Object Detection (AAAI 2024 · similarity 0.83)
- A New Adversarial Perspective for LiDAR-based 3D Object Detection (AAAI 2025 · similarity 0.83)
- SwiftPillars: High-Efficiency Pillar Encoder for Lidar-Based 3D Detection (AAAI 2024 · 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.
