Revisiting Noise Resilience Strategies in Gesture Recognition: Short-Term Enhancement in sEMG Analysis
Abstract: Gesture recognition based on surface electromyography (sEMG) has been gaining importance in many 3D Interactive Scenes. However, sEMG is easily influenced by various forms of noise in real-world environments, leading to challenges in providing long-term stable interactions through sEMG. Existing methods often struggle to enhance model noise resilience through various predefined data augmentation techniques.
In this work, we revisit the problem from a short-term enhancement perspective to improve precision and robustness against various common noisy scenarios with learnable denoise using sEMG intrinsic pattern information and sliding-window attention. We propose a Short Term Enhancement Module(STEM), which can be easily integrated with various models. STEM offers several benefits: 1) Noise-resistant, enhanced robustness against noise without manual data augmentation; 2) Adaptability, adaptable to various models; and 3) Inference efficiency, achieving short-term enhancement through minimal weight-sharing in an efficient attention mechanism.
In particular, we incorporate STEM into a transformer, creating the Short-Term Enhanced Transformer (STET).
Compared with best-competing approaches, the impact of noise on STET is reduced by more than 20\%. We report promising results on classification and regression tasks and demonstrate that STEM generalizes across different gesture recognition tasks. The code is available at https://anonymous.4open.science/r/short_term_semg.
Lay Summary: Gesture recognition technology, which translates human movements into digital commands, is growing increasingly popular in virtual reality, robotics, and assistive devices. This technology often uses electrical signals from muscles (known as sEMG signals). However, these signals can easily become noisy or unreliable due to environmental factors like skin moisture or poor electrode contact, making accurate recognition difficult.
In our work, we introduce a new approach to help computers better interpret these noisy signals. We created a simple yet effective technique called STEM that enhances short-term signal features, making gesture recognition significantly more reliable without needing extra data. This method can be effortlessly added to existing recognition systems. When tested, our approach showed greater accuracy and was notably less affected by noise than current methods. This innovation could substantially improve how reliably we interact with computers and robots in everyday life, especially in environments that are less controlled than laboratories.
Application-Driven Machine Learning: This submission is on Application-Driven Machine Learning.
Primary Area: Applications->Neuroscience, Cognitive Science
Keywords: Gesture Recognition, Human Brain Interaction, sEMG, Short Term Enhanced Transformer
Submission Number: 7220
Loading