Deep Learning for Skeleton Based Human Motion Rehabilitation Assessment: A Benchmark

2025-07-28

AI Summary

Standardized benchmarks and reproducible automated motion assessment methods are currently lacking in rehabilitation research, hindering methodological progress and cross-study comparability. To address this, we focus on skeleton-based movement quality assessment and introduce Rehab-Pile—the first open-source, unified benchmark for rehabilitation motion analysis. It encompasses multi-scenario, multi-task (classification and regression) skeletal motion data collected from clinical and home-based settings. We design a general-purpose evaluation framework that integrates video-based skeleton extraction with diverse deep learning architectures, enabling fair, standardized comparison across methods. We conduct large-scale benchmarking experiments and publicly release all data, code, and results. This work fills a critical gap in standardized evaluation for rehabilitation motion assessment, significantly enhancing reproducibility, transparency, and practical applicability. By establishing a rigorous, accessible foundation, it advances the development of low-cost, personalized digital rehabilitation technologies.

Technology Category

• Intelligent Robots: Motion and Path Planning
• Computer Vision: Motion & Tracking
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Automated assessment of human motion plays a vital role in rehabilitation, enabling objective evaluation of patient performance and progress. Unlike general human activity recognition, rehabilitation motion assessment focuses on analyzing the quality of movement within the same action class, requiring the detection of subtle deviations from ideal motion. Recent advances in deep learning and video-based skeleton extraction have opened new possibilities for accessible, scalable motion assessment using affordable devices such as smartphones or webcams. However, the field lacks standardized benchmarks, consistent evaluation protocols, and reproducible methodologies, limiting progress and comparability across studies. In this work, we address these gaps by (i) aggregating existing rehabilitation datasets into a unified archive called Rehab-Pile, (ii) proposing a general benchmarking framework for evaluating deep learning methods in this domain, and (iii) conducting extensive benchmarking of multiple architectures across classification and regression tasks. All datasets and implementations are released to the community to support transparency and reproducibility. This paper aims to establish a solid foundation for future research in automated rehabilitation assessment and foster the development of reliable, accessible, and personalized rehabilitation solutions. The datasets, source-code and results of this article are all publicly available.

Problem

• Lack of standardized benchmarks for rehabilitation motion assessment
• Need for consistent evaluation protocols in deep learning methods
• Absence of reproducible methodologies in rehabilitation assessment research

Innovation

• Aggregated datasets into unified Rehab-Pile archive
• Proposed benchmarking framework for deep learning
• Benchmarked architectures across multiple tasks