Video motion capture showed uneven accuracy across movement measurements
A review of Theia3D found strong agreement for many walking measures, while hip bending angles and some joint rotations were less accurate.
Based on the published abstract. The full paper may contain additional methods, results and limitations.
The 30-second takeaway
The reviewed video system showed strong agreement with reference methods for many walking measures, but poorer accuracy for some joint angles. A favorable finding for one measurement does not establish accuracy for every output. These results support examining the specific measurement and task; the abstract leaves performance across broader activities and populations uncertain.
Most studies focused on walking or running. The authors call for better comparison methods and standardized biomechanical models, so the findings cannot establish consistent performance across all functional tasks or population groups.
What the review evaluated
The researchers assessed whether markerless video measurements agreed with reference methods and stayed consistent across sessions. Most studies tested movement measurements during walking or running, leaving limited coverage of other functional tasks in the abstract.
Accuracy varied by measurement
Many walking measures had strong agreement with reference methods and good repeatability. Overall lower-limb movement measurements showed reasonable accuracy, but hip bending and straightening angles and rotations in the horizontal plane were less accurate.
The original publication
The accuracy, validity and reliability of Theia3D markerless motion capture for studying the biomechanics of human movement: A systematic review.
Varcin F, Boocock MG
Artif Intell Med · 2026
- PubMed ID
- 41518893
- Record checked
AI-assisted research and writing. This explains one selected publication; it is not a complete review of everything known. Our approach.
One more question, understood.
Keep track of the research you’ve explored.