Self-supervised learning enhances accuracy and data efficiency in lower-limb joint moment estimation from gait kinematics

View Full Paper

Ask AI

Bookmark

Authors

Discussion

Experimental analysis reveals improved estimation of joint moments in children using self-supervised learning, indicating greater data efficiency.

The fine-tuned self-supervised learning model significantly outperformed the baseline model in joint moment estimation.
Mean Squared Errors (MSEs) reduced from 24.00% to 45.16%, demonstrating strong improvements in data efficiency.
A Transformer auto-encoder was used to pre-train the model on unlabeled data, effectively capturing human motion features.
The self-supervised learning method reduces data collection burdens, making it beneficial for clinical applications.

Li et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark

Experimental analysis reveals improved estimation of joint moments in children using self-supervised learning, indicating greater data efficiency.