Abstract
Surface electromyography (sEMG) is a low-power, cost-effective biosignal for hand-pose estimation and gesture classification. In this work, we examine whether self-supervised pretraining on sEMG can yield transferable representations for continuous hand-pose estimation. We introduce EMG-GPT, a causal transformer-based model that operates on discrete sEMG representations from a frozen residual vector quantization (RVQ) tokenizer and learns temporal dynamics through depth-autoregressive future-code prediction. The model combines within-frame integration with causal temporal modeling while preserving the geometry of the pretrained codebook. EMG-GPT shows competitive results in both Regression and Tracking tasks, supporting EMG-only pretraining as a viable approach for learning transferable sEMG representations.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Magni, E., Potamias, R. A., Zafeiriou, S., & Barmpas, K. (2026). EMG-GPT: Predictive Pretraining on Residual-Quantized EMG Tokens for Hand Pose Estimation. https://omanscience.com/en/articles/emg-gpt-predictive-pretraining-on-residual-quantized-emg-tokens-for-hand-pose-estimation
MLA 9
Magni, Ettore, et al. "EMG-GPT: Predictive Pretraining on Residual-Quantized EMG Tokens for Hand Pose Estimation." https://omanscience.com/en/articles/emg-gpt-predictive-pretraining-on-residual-quantized-emg-tokens-for-hand-pose-estimation.
Chicago (author–date)
Magni, Ettore, Rolandos Alexandros Potamias, Stefanos Zafeiriou, and Konstantinos Barmpas. 2026. "EMG-GPT: Predictive Pretraining on Residual-Quantized EMG Tokens for Hand Pose Estimation." https://omanscience.com/en/articles/emg-gpt-predictive-pretraining-on-residual-quantized-emg-tokens-for-hand-pose-estimation.
Harvard
Magni, E., Potamias, R. A., Zafeiriou, S. and Barmpas, K. (2026) 'EMG-GPT: Predictive Pretraining on Residual-Quantized EMG Tokens for Hand Pose Estimation', Available at: https://omanscience.com/en/articles/emg-gpt-predictive-pretraining-on-residual-quantized-emg-tokens-for-hand-pose-estimation.
Vancouver
Magni E, Potamias RA, Zafeiriou S, Barmpas K. EMG-GPT: Predictive Pretraining on Residual-Quantized EMG Tokens for Hand Pose Estimation. https://omanscience.com/en/articles/emg-gpt-predictive-pretraining-on-residual-quantized-emg-tokens-for-hand-pose-estimation
IEEE
E. Magni, R. A. Potamias, S. Zafeiriou, and K. Barmpas, "EMG-GPT: Predictive Pretraining on Residual-Quantized EMG Tokens for Hand Pose Estimation," https://omanscience.com/en/articles/emg-gpt-predictive-pretraining-on-residual-quantized-emg-tokens-for-hand-pose-estimation.