发表机构
Indian Institute of Technology Kharagpur; Indiana University Indianapolis; Bangladesh University of Engineering and Technology; University of Toronto（印度理工学院卡哈拉格普尔分校; 印第安纳大学印第安纳波利斯分校; 孟加拉国工程技术大学; 多伦多大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究提出基于膜电位统计的低计算量OOD检测方法MDD，并构建事件相机损坏基准Gen1-C，在最高损坏下对六种损坏中的五种实现AUROC超过0.88，验证了膜电位动力学在SNN事件感知中的有效性。
AI 中文摘要
脉冲神经网络（SNNs）为处理事件相机数据提供了一种节能的方法，然而在此设置下，分布外（OOD）检测仍然具有挑战性。现有的OOD检测方法通常依赖于模型输出或计算组件，而这些在目标检测SNNs中不可用，或者不适合低计算量部署。为此，我们展示了阈下膜电位 \\(V_{\mathrm{mem}}(t)\\) 为检测分布偏移提供了有用的内部信号。从这些膜动力学中导出的简单逐通道统计量能够实现OOD检测。为了评估这种方法，我们引入了Gen1-C，这是一个基于Prophesee Gen1汽车检测数据集构建的事件相机损坏基准，包含六个受传感器启发的直方图级压力测试，每个测试有五个严重级别。我们进一步提出了多描述符偏差（MDD），这是一种对损坏不敏感的方法，作用于膜电位统计量。在最高损坏严重度下，MDD仅使用有界的64帧观测窗口，在六种损坏中的五种上实现了超过0.88的AUROC。值得注意的是，剩余的损坏也是对所基于检测器影响最小的那一种。这些结果表明，时间膜电位动力学可以为基于SNN的事件感知中的OOD检测提供有效且低成本的信号。
英文摘要
Spiking Neural Networks (SNNs) offer an energy-efficient approach to processing event-camera data, yet out-of-distribution (OOD) detection remains challenging in this setting. Existing OOD detection methods often depend on model outputs or computational components that are unavailable in object detection SNNs or are poorly suited to low-compute deployment. To that effect, we show that the subthreshold membrane potential \(V_{\mathrm{mem}}(t)\) provides a useful internal signal for detecting distribution shifts. Simple per-channel statistics derived from these membrane dynamics enable OOD detection. To evaluate this approach, we introduce Gen1-C, an event-camera corruption benchmark developed upon the Prophesee Gen1 automotive detection dataset, containing six sensor-motivated histogram-level stress tests at five severity levels. We further propose the Multi-Descriptor Deviation (MDD), a corruption-blind method that operates on membrane-potential statistics. At the highest corruption severity, MDD achieves an AUROC of more than 0.88 on five of the six corruptions using only a bounded 64-frame observation window. Notably, the remaining corruption is also the one that has the smallest effect on the underlying detector. These results show that the temporal membrane-potential dynamics can provide an effective and low-cost signal for OOD detection in SNN-based event perception.
