arXiv：cs.LG· Tianhao Wu, Xu Wu, Amirmohammad Radmehr, Jiawei Yu, Yi Wu, Phuc Nguyen, Jian Liu·· 3 小时前AI 评分34
LiteEMG-FM：面向鲁棒 EMG 传感的高效可部署基础模型
LiteEMG-FM: An Efficient and Deployable Foundation Model for Robust EMG Sensing
AI 导读
LiteEMG-FM 是一个混合 CNN-Transformer 基础模型，在 16 个上下肢 EMG 数据集上预训练，可跨用户与数据集泛化。它采用分层唤醒架构，由常驻轻量 1D-CNN 过滤静息与非目标活动，仅在有效手势时激活主模型，并评估了完全卸载、拆分推理与全端侧处理三种部署方式。在零校准跨参与者与数据稀缺条件下，其表现优于 SOTA 时间序列基础模型和监督基线。
正文
Abstract:Electromyography (EMG) signals vary substantially across individuals, body regions, recording sessions, and sensing hardware, limiting the generalization of models for assistive devices and human-computer interaction. Existing time-series foundation models are also computationally expensive for real-time wearable deployment and often fail to capture EMG-specific time-frequency characteristics. We present LiteEMG-FM, an efficient hybrid CNN-Transformer foundation model for practical EMG sensing. Pretrained on 16 diverse upper- and lower-limb EMG datasets, LiteEMG-FM learns representations that generalize across users and datasets. For resource-constrained deployment, we implement a hierarchical wake-up architecture in which a lightweight, always-on 1D-CNN filters rest and non-target activity and activates LiteEMG-FM only for valid gestures. We evaluate full inference offloading, split inference, and full on-device processing, characterizing their trade-offs in latency, power consumption, and memory footprint. Across diverse evaluation settings, LiteEMG-FM outperforms state-of-the-art time-series foundation models and supervised baselines, particularly under zero-calibration cross-participant and data-scarce conditions. These results demonstrate that LiteEMG-FM is an effective, efficient, and deployable foundation model for EMG applications.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.02497 [cs.LG] |
| (or arXiv:2610.02497v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.02497 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Tianhao Wu [view email]
[v1]
Thu, 1 Oct 2026 21:19:51 UTC (1,415 KB)
来源：arXiv：cs.LG · arxiv.org