让智能手表本地实时识别25种以上动作，速度快且保护隐私。
WatchHAR: Real-time On-device Human Activity Recognition System for Smartwatches
- 端到端训练模块融合音频与惯性数据预处理和推理
- 动作事件检测仅需9.3毫秒，分类11.8毫秒，准确率超90%
- 适合对隐私和低延迟有要求的可穿戴设备应用
尽管在实用且多模态细粒度人体活动识别方面取得进展，但在非受限环境下完全在智能手表上运行的系统仍难以实现。我们提出WatchHAR，一种基于音频和惯性数据的端到端活动识别系统，可在智能手表上独立运行，解决外部数据处理带来的隐私和延迟问题。通过优化整个流程中的每个组件，WatchHAR实现累积性能提升。我们引入一种新型架构，将传感器数据预处理与推理统一为端到端可训练模块，在保持超过25种活动类别90%以上准确率的同时，实现5倍加速。WatchHAR在事件检测和活动分类任务上优于现有最先进模型，直接在手表上运行时，活动事件检测耗时9.3毫秒，多模态活动分类耗时11.8毫秒。该研究推动了本地化活动识别的发展，使智能手表真正成为独立、隐私友好且侵入性极小的持续活动追踪设备。
原文摘要 · Abstract (English)
Despite advances in practical and multimodal fine-grained Human Activity Recognition (HAR), a system that runs entirely on smartwatches in unconstrained environments remains elusive. We present WatchHAR, an audio and inertial-based HAR system that operates fully on smartwatches, addressing privacy and latency issues associated with external data processing. By optimizing each component of the pipeline, WatchHAR achieves compounding performance gains. We introduce a novel architecture that unifies sensor data preprocessing and inference into an end-to-end trainable module, achieving 5x faster processing while maintaining over 90% accuracy across more than 25 activity classes. WatchHAR outperforms state-of-the-art models for event detection and activity classification while running directly on the smartwatch, achieving 9.3 ms processing time for activity event detection and 11.8 ms for multimodal activity classification. This research advances on-device activity recognition, realizing smartwatches' potential as standalone, privacy-aware, and minimally-invasive continuous activity tracking devices.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。