Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces GenGait, a Transformer-based masked autoencoder trained on normative gait data for anomaly detection and kinematic correction in human gait analysis. The model estimates joint inconsistency scores by occluding individual joints and measuring deviations from the learned normative prior, then reconstructs the full skeleton with flagged joints withheld to yield corrected kinematic trajectories. Experiments on simulated gait abnormalities demonstrate accurate localization of inconsistent joints and significant reduction in angular deviation, enabling interpretable, subject-specific impairment localization without disease labels.
Spot and fix gait abnormalities without needing labeled disease data: GenGait uses a Transformer to learn normal movement, then flags and corrects deviations.
Gait analysis provides an objective characterization of locomotor function and is widely used to support diagnosis and rehabilitation monitoring across neurological and orthopedic disorders. Deep learning has been increasingly applied to this domain, yet most approaches rely on supervised classifiers trained on disease-labeled data, limiting generalization to heterogeneous pathological presentations. This work proposes a label-free framework for joint-level anomaly detection and kinematic correction based on a Transformer masked autoencoder trained exclusively on normative gait sequences from 150 adults, acquired with a markerless multi-camera motion-capture system. At inference, a two-pass procedure is applied to potentially pathological input sequences, first it estimates joint inconsistency scores by occluding individual joints and measuring deviations from the learned normative prior. Then, it withholds the flagged joints from the encoder input and reconstructs the full skeleton from the remaining spatiotemporal context, yielding corrected kinematic trajectories at the flagged positions. Validation on 10 held-out normative participants, who mimicked seven simulated gait abnormalities, showed accurate localization of biomechanically inconsistent joints, a significant reduction in angular deviation across all analyzed joints with large effect sizes, and preservation of normative kinematics. The proposed approach enables interpretable, subject-specific localization of gait impairments without requiring disease labels. Video is available at https://youtu.be/Rcm3jqR5pN4.