AI motion capture system for enhancing human motor function
Project overview
Project title
AI-driven markerless motion capture system for enhancing human motor function screening and assessment
Project description
This project aims to advance the field of human movement science by addressing the challenges encountered when developing a low-cost, automated system for screening the movement of pre-elite student-athletes. Leveraging state-of-the-art artificial intelligence (AI), markerless motion capture and stereo vision technologies, this research will tackle critical challenges in biomechanics and sports science.
The overarching vision is that the developed system will integrate sport-specific performance metrics with biomechanical analyses, enabling deeper insights into motor function and movement strategies. This innovative approach has the potential to revolutionize athlete assessment, providing actionable feedback for injury prevention, performance enhancement, and tailored training programs.
By contributing to the fields of AI, biomechanics, and applied sports science, this project offers a unique opportunity for PhD candidates to engage in interdisciplinary research with real-world impact.
Supervisory team
University
| Name of university supervisor | Dr Ben Hoffman, Dr Patricio Pincheira, Prof Stephen Bird |
| Name of university | University of Southern Queensland |
| Email address | ben.hoffman@unisq.edu.au |
| Faculty | School of Health & Medical Sciences |
CSIRO
| Name of CSIRO supervisor | Dr David Ahmedt Aristizabal |
| Email address | david.ahmedtaristizabal@data61.csiro.au |
| CSIRO Research Unit | Data 61 |
Industry
| Name of industry supervisor | Jackson Stone |
| Name of business/organisation | Toowoomba Grammar School |
| Email address | j.stone@twgs.qld.edu.au |
Further details
| Industry partner facilities and support (Toowoomba Grammar School) | Technology & Training: Hands-on access to GPS units, MyVert technology, and expert training to support the development of targeted training analysis. Professional Development: Weekly education sessions led by industry leaders in strength and conditioning, sport science, injury management, and sport-specific training. World-Class Facilities: Engage in your research within state-of-the-art education facilities and premier sports training environments. |
| Primary location of student | University of Southern Queensland, 11 Salisbury Road, Ipswich QLD 4305 |
| Industry engagement component location | Toowoomba Grammar School, 24 Margaret Street, East Toowoomba QLD 4350 |
| Other locations | CSIRO Black Mountain, 2-40 Clunies Ross Street, Acton, ACT 2601 |
| Ideal student skillset | Essential Skills: Bachelor’s degree with First Class Honours or Second Class Honours (Division A) or equivalent (thesis comprising at least two units) or Master’s degree (thesis comprising at least two units) or equivalent, in computer science, biomechanics, sports science, or a related field. Basic understanding of machine learning algorithms and deep learning models. Proficiency in Python in a Linux environment and development experience using Tensorflow or PyTorch. Strong linear algebra and computer vision knowledge. Demonstrated responsibility, perseverance, and commitment to achieving project goals and meeting deadlines. Desirable Skills: Experience with biomechanical analysis; sports science research or athlete performance assessment. Familiarity with stereo camera systems. Ability to work collaboratively in an interdisciplinary team. Strong communication skills. |
| Application Close Date | Open until position filled |
| Apply | UniSQ |