TEMPEST：基于角度间隔学习的时间嵌入实现可扩展驾驶员识别
原标题：TEMPEST: Temporal Embeddings for Scalable Driver Identification via Angular Margin Learning
AI 摘要
研究者提出 TEMPEST，一种基于时间卷积网络并用 ArcFace 加性角度间隔损失训练的驾驶行为嵌入模型，将 60 秒多模态驾驶窗口映射为 96 维嵌入。在 45 名驾驶员的 DriverID 数据集上，其 Rank-1 准确率达 91.71%，比最佳经典模型高 17.9 个百分点，比最强 triplet-loss 基线高 58.4 个百分点；驾驶员池从 10 增至 45 时准确率仅下降 4.3 个百分点。模型仅 720K 参数、2.80 MB，支持无需重训练的动态注册，并在公开 KIA Soul 数据集上验证了跨会话优势。
正文节选
TEMPEST: Temporal Embeddings for Scalable Driver Identification via Angular Margin Learning Abstract Scalable driver identification requires embedding models that maintain discriminative performance as fleet size grows, yet existing triplet-loss formulations degrade rapidly with driver pool size and overfit to session-specific patterns under rigorous temporal evaluation. We introduce TEMPEST, a Temporal Convolutional Network embedding model trained with an additive angular margin (ArcFace) loss