AgentNorm：面向相机协同感知的 Agent 级测试时归一化方法
Agent-Wise Test-Time Normalization for Camera-Based Cooperative Perception with Corrupted Collaborators
AI 导读
arXiv 论文提出 AgentNorm，一种针对相机协同感知中协作者退化的 agent 级测试时归一化方法。传统混合统计归一化会使干净自车检测下降 3.58–6.68 [email protected]，AgentNorm 让每个协作者用自身观测估计统计量，并保持自车路径不变。
正文 · 原文
Abstract:Intermediate-fusion cooperative perception lets a vehicle detect objects beyond its own field of view by fusing features shared by nearby agents. When only a collaborator's cameras are degraded by fog, motion blur or exposure errors, the features it shares drift away from the training distribution while the ego input stays clean, and cooperation can become worse than not cooperating at all. Test-time batch-normalization adaptation, the standard training-free remedy, estimates statistics over the pooled batch of all agents and therefore normalizes both the clean ego and the corrupted collaborator with a mixture of two distributions. We show that this mixing harms even the clean ego: when the ego shares BN statistics with a corrupted collaborator, its detection from its own features drops by 3.58 to 6.68 [email protected] points although its cameras are untouched. We propose AgentNorm, an agent-wise test-time normalization that i) estimates the normalization statistics of every collaborator from its own current observation and ii) anchors the ego to its source statistics, so that the ego path remains exactly the original model. AgentNorm has no learnable parameters, needs no gradients or labels, is independent of the order of test frames, and runs locally on each agent without changing the transmitted features. On camera-based OPV2V with four fusion checkpoints and collaborator-only corruptions, AgentNorm raises [email protected] under severity-5 fog from 29.42-37.25 to 43.32-58.86 and outperforms pooled normalization by 5.53-12.94 points under fog and motion blur, at an extra latency of 2.3 ms per frame. On the real-world DAIR-V2X-C dataset, it outperforms pooled normalization in all ten infrastructure-corruption settings. We also report the cases it does not fix: occlusion, which removes image content rather than shifting feature statistics, and the case in which all agents share the same shift.
| Comments: | 11 pages, 8 figures |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.39505 [cs.CV] |
| (or arXiv:2610.39505v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.39505 arXiv-issued DOI via DataCite |
Submission history
From: Yu Wang [view email]
[v1]
Thu, 29 Oct 2026 07:05:47 UTC (809 KB)
来源：arXiv cs.AI · arxiv.org