Subject-independent gait activity recognition using DSAF: dual-stream IMU-EMG attention fusion with…

StudyFrontiers in neurorobotics· 2026

Subject-independent gait activity recognition using DSAF: dual-stream IMU-EMG attention fusion with asymmetric temporal encoding and physiological complementarity weighting.

Zhangyue Xu, Yan Zhu, Shurui Li

Assistive Technology & WearablesStudyQ2IF 3.6

Dual-stream IMU-EMG fusion model accurately classifies gait activities in unseen subjects.

Subject-independent gait activity recognition using DSAF: dual-stream IMU-EMG attention fusion with asymmetric temporal encoding and physiological complementarity weighting.

Zhangyue Xu … Shurui Li

Frontiers in neurorobotics · 2026

Background

Accurate gait recognition using wearable sensors is of significant clinical value for adaptive prosthetic control, lower-limb exoskeleton assistance, and objective rehabilitation assessment.

Methods

To address these gaps, this study proposes DSAF, a dual-stream attention fusion network that separately encodes kinematic (IMU) and neuromuscular (EMG) information, and adaptively integrates them through a physiological complementarity weighting mechanism designed for window-level modality adaptation.

n = 18 held-out subjects

Results

correctly identified gait activity in new users almost every time

Accuracy

96.41%

Macro-precision

96.08%

Macro-recall

95.62%

Macro-F1

95.81%

More results

Ablation studies further confirm that both the modality-specific encoding and the adaptive fusion mechanism contribute positively to the performance.

“

Conclusion

DISCUSSION: These findings indicate that adaptive IMU-EMG fusion can effectively strengthen wearable gait recognition, providing a promising solution for real-world rehabilitation monitoring and assistive human-machine interfaces.

Subject-independent gait activity recognition using DSAF: dual-stream IMU-EMG attention fusion with asymmetric temporal encoding and physiological complementarity weighting.

Subject-independent gait activity recognition using DSAF: dual-stream IMU-EMG attention fusion with asymmetric temporal encoding and physiological complementarity weighting.

Related papers