EFormer：用于连续sEMG手部姿态跟踪的时间对齐局部校正
EFormer: Temporally Aligned Local Correction for Continuous sEMG-Based Hand Pose Tracking
浏览论文内容
中文总结 AI 辅助
EFormer提出一种基于冻结跟踪骨干的残差特征校正网络，融合高速事件分支与时间对齐交叉注意力，在连续sEMG手部姿态跟踪中降低MAE 11.38%。
中文摘要 AI 辅助
表面肌电信号（sEMG）提供了一种可穿戴、无摄像头的连续手部运动推断信号。由于记录的波形是间接测量，其与运动的关系随时间变化，且个体解剖结构和传感器放置位置会改变信号分布，因此将肌肉活动映射到关节运动学仍然具有挑战性。本文提出了EFormer，一种基于冻结跟踪骨干的残差特征校正网络。EFormer结合了高速事件分支、时间对齐的局部交叉注意力、两个因果旋转位置编码（RoPE）时间层以及一个有界、动态门控的残差。EFormer接收以2 kHz采样的16通道sEMG，并将25 Hz下的64通道跟踪表示与200 Hz下的128通道事件表示融合。交叉注意力使用100 ms的名义延迟、300 ms的历史参数和50 ms的容差；其因果掩码将每个查询限制为发生在50-400 ms之前的事件。校正比例为0.15。所评估的延续训练配置包含585,376个可训练参数和5,974,508个冻结参数。在测试集上，EFormer实现了0.1546634 rad的MAE、0.24063 rad的RMSE和0.74801的R²，而官方跟踪基线的对应值为0.1745326 rad、0.2715448 rad和0.6791103。EFormer相对于基线将MAE降低了11.38%。结果表明，时间对齐的事件特征校正可以减少连续手部姿态跟踪误差。
英文摘要
Surface electromyography (sEMG) provides a wearable, camera-free signal for continuous hand-motion inference. Mapping muscle activity to joint kinematics remains challenging because the recorded waveforms are indirect measurements, their relationship with motion changes over time, and individual anatomy and sensor placement alter the signal distribution. This paper presents EFormer, a residual feature-correction network built on a frozen tracking backbone. EFormer combines a high-rate event branch, temporally aligned local cross-attention, two causal rotary position embedding (RoPE) temporal layers, and a bounded, dynamically gated residual. EFormer receives 16-channel sEMG sampled at 2 kHz and fuses a 64-channel tracking representation at 25 Hz with a 128-channel event representation at 200 Hz. Cross-attention uses a nominal delay of 100 ms, a 300 ms history parameter, and a 50 ms tolerance; its causal mask restricts each query to events occurring 50-400 ms earlier. The correction scale is 0.15. The evaluated continuation-training configuration contains 585,376 trainable parameters and 5,974,508 frozen parameters. On the test set, EFormer achieves an MAE of 0.1546634 rad, an RMSE of 0.24063 rad, and an R^2 of 0.74801, compared with 0.1745326 rad, 0.2715448 rad, and 0.6791103 for the official tracking baseline. EFormer reduces MAE by 11.38% relative to the baseline. The results show that temporally aligned event-feature correction can reduce continuous hand-pose tracking error.
