Research map / Detection and segmentation / 3D detection and driving
Detecting Temporal Misalignment Attacks in Multimodal Fusion for Autonomous Driving
Md Hasan Shahriar, Md Mohaimin Al Barat, Harshavardhan Sundar, Ning Zhang, Naren Ramakrishnan, Thomas Hou, Wenjing Lou
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 |
|---|---|---|---|
| attack both camera and LiDAR with a constant freeze instead of one stream | AION · KITTI | AUROC: 0.9717 -> 0.1274 | Hurt |
| attack both camera and LiDAR with reversal instead of one stream | AION · nuScenes | AUROC: 0.9554 -> 0.5095 | Hurt |
| attack both streams with random replacement instead of camera only | AION · KITTI | AUROC: 0.9214 -> 0.9092 | Mixed |
| use larger window size w >= 7 instead of w = 3 or 5 (ablation) | AION · KITTI and nuScenes | AUROC | Hurt |
| use AION instead of timestamp-sanity baseline | AION · KITTI and nuScenes | AUROC: 0.85-0.97 vs often below 0.8 on KITTI and 0.7 on nuScenes | Improved |
| use AION instead of sliding-window correlation baseline | AION · nuScenes | AUROC | Improved |
| calibrate threshold at 99th percentile on benign data (window-level TPR at FPR < 0.01) | AION · KITTI | TPR: 1.0000 for Reversal both vs 0.0600 for Constant both | Mixed |
| run AION alongside perception model as plug-in | BEVFusion · nuScenes | latency: 3.26 ms | No clear effect |
Closest papers by meaning
- E2E-MFD: Towards End-to-End Synchronous Multimodal Fusion Detection (NeurIPS 2024 · similarity 0.84)
- BEVSync: Asynchronous Data Alignment for Camera-based Vehicle-Infrastructure Cooperative Perception Under Uncertain Delays (AAAI 2025 · similarity 0.84)
- COMMIT: Certifying Robustness of Multi-Sensor Fusion Systems Against Semantic Attacks (AAAI 2025 · similarity 0.84)
- Fusion Is Not Enough: Single Modal Attacks on Fusion Models for 3D Object Detection (ICLR 2024 · similarity 0.84)
- SDAC: A Multimodal Synthetic Dataset for Anomaly and Corner Case Detection in Autonomous Driving (AAAI 2024 · similarity 0.83)
- AlterMOMA: Fusion Redundancy Pruning for Camera-LiDAR Fusion Models with Alternative Modality Masking (NeurIPS 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.
