mlsanityData to 14 Oct 2026

A Standardized Benchmark for Skeleton-Based Rehabilitation Assessment Using Deep Learning

arXiv 2507.21018 ↗First posted 28 Jul 2025cs.CV cs.LGcited by 2

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 18 of its 42 references that are in the map. First posted on arXiv 28 Jul 2025.

Builds on

18 of 42 references.

References · 18

  1. 2017Attention Is All You Needcited by 60,361
  2. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
  3. 2016Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learningcited by 1,512
  4. 2018Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognitioncited by 1,209
  5. 2018OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fieldscited by 1,162
  6. 2021Action-Conditioned 3D Human Motion Synthesis with Transformer VAEcited by 501
  7. 2020Action2Motion: Conditioned Generation of 3D Human Motionscited by 441
  8. 2019InceptionTime: Finding AlexNet for Time Series Classificationcited by 387
  9. 2021MotionRNN: A Flexible Model for Video Prediction with Spacetime-Varying Motionscited by 54
  10. 2024SkateFormer: Skeletal-Temporal Transformer for Human Action Recognitioncited by 47
  11. 2024aeon: a Python toolkit for learning from time seriescited by 47
  12. 2023Improving Position Encoding of Transformers for Multivariate Time Series Classificationcited by 45
6 more
  1. 2023An Approach to Multiple Comparison Benchmark Evaluations that is Stable Under Manipulation of the Comparate Setcited by 29
  2. 2019A Deep Learning Framework for Assessing Physical Rehabilitation Exercisescited by 26
  3. 2020A Review of Computational Approaches for Evaluation of Rehabilitation Exercisescited by 9
  4. 2024Look Into the LITE in Deep Learning for Time Series Classificationcited by 9
  5. 2024A Medical Low-Back Pain Physical Rehabilitation Dataset for Human Body Movement Analysiscited by 5
  6. 2023An Examination of Wearable Sensors and Video Data Capture for Human Exercise Classificationcited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026The Multiverse of Time Series Machine Learning: an Archive for Multivariate Time Series Classificationcited by 4
  2. 2026When Does Exercise-Specific Joint Selection Help? An Audit of Evaluation and Control Designcited by 0