BioGait-VLM: A Tri-Modal Vision-Language-Biomechanics Framework for Interpretable Clinical Gait Assessment

TL;DR

BioGait-VLM introduces a tri-modal framework combining vision, language, and biomechanics for interpretable and robust clinical gait assessment, outperforming existing methods and enhancing clinical plausibility.

Contribution

The paper presents a novel tri-modal architecture with biomechanical tokenization and a new benchmark dataset for improved, interpretable gait analysis.

Findings

Achieves state-of-the-art recognition accuracy on an expanded dataset.

Biomechanical tokens improve clinical plausibility and evidence grounding.

Framework generalizes better by capturing rhythmic dynamics and joint mechanics.

Abstract

Video-based Clinical Gait Analysis often suffers from poor generalization as models overfit environmental biases instead of capturing pathological motion. To address this, we propose BioGait-VLM, a tri-modal Vision-Language-Biomechanics framework for interpretable clinical gait assessment. Unlike standard video encoders, our architecture incorporates a Temporal Evidence Distillation branch to capture rhythmic dynamics and a Biomechanical Tokenization branch that projects 3D skeleton sequences into language-aligned semantic tokens. This enables the model to explicitly reason about joint mechanics independent of visual shortcuts. To ensure rigorous benchmarking, we augment the public GAVD dataset with a high-fidelity Degenerative Cervical Myelopathy (DCM) cohort to form a unified 8-class taxonomy, establishing a strict subject-disjoint protocol to prevent data leakage. Under this setting, BioGait-VLM achieves state-of-the-art recognition accuracy. Furthermore, a blinded expert study confirms that biomechanical tokens significantly improve clinical plausibility and evidence grounding, offering a path toward transparent, privacy-enhanced gait assessment.