TinierHAR在边缘设备上实现超轻量人体活动识别，精度不降反升。
TinierHAR: Towards Ultra-Lightweight Deep Learning Models for Efficient Human Activity Recognition on Edge Devices
- 融合残差深度可分离卷积与门控循环单元，提升空间-时序建模效率。
- 参数量减少2.7倍（比TinyHAR）、计算量降低6.4倍，平均F1分数持平。
- 首次系统拆解各组件贡献，提供可复用的轻量化设计指南。
资源受限的可穿戴设备对高效的人体活动识别（HAR）模型提出挑战。本文提出TinierHAR，一种超轻量级深度学习架构，通过结合残差深度可分离卷积、门控循环单元（GRUs）和时序聚合，在不牺牲性能的前提下实现最优效率。在14个公开HAR数据集上的评估显示，相比TinyHAR，TinierHAR参数量减少2.7倍，乘加操作（MACs）减少6.4倍；相比DeepConvLSTM，参数量减少43.3倍，MACs减少58.6倍，同时保持平均F1分数稳定。本工作还首次开展系统性消融实验，解析TinierHAR、SOTA模型TinyHAR及经典DeepConvLSTM中空间-时序模块的贡献，为高效HAR系统设计提供可行动洞察。最后，研究总结发现并提出原则性设计建议。为推动边缘HAR研究，所有代码与材料已开源。
原文摘要 · Abstract (English)
Human Activity Recognition (HAR) on resource-constrained wearable devices demands inference models that harmonize accuracy with computational efficiency. This paper introduces TinierHAR, an ultra-lightweight deep learning architecture that synergizes residual depthwise separable convolutions, gated recurrent units (GRUs), and temporal aggregation to achieve SOTA efficiency without compromising performance. Evaluated across 14 public HAR datasets, TinierHAR reduces Parameters by 2.7x (vs. TinyHAR) and 43.3x (vs. DeepConvLSTM), and MACs by 6.4x and 58.6x, respectively, while maintaining the averaged F1-scores. Beyond quantitative gains, this work provides the first systematic ablation study dissecting the contributions of spatial-temporal components across proposed TinierHAR, prior SOTA TinyHAR, and the classical DeepConvLSTM, offering actionable insights for designing efficient HAR systems. We finally discussed the findings and suggested principled design guidelines for future efficient HAR. To catalyze edge-HAR research, we open-source all materials in this work for future benchmarking\footnote{https://github.com/zhaxidele/TinierHAR}
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。