ÌÇÐÄTV

The project: Human gait analysis: a pilot study using low-cost single monocular video-based motion capture (MoCap)

This project explored two low-cost, scalable alternatives for markerless motion capture (MoCap): monocular camera-based MoCap and mm-Wave sensing. Both approaches offer simple setup, longer-duration recordings and reduced hardware requirements compared with conventional expensive systems. Monocular cameras provide portable and unobtrusive deployment, while mm-Wave sensing offers privacy-preserving and lighting-independent motion capture.

A pilot study with healthy participants assessed the technical feasibility of both approaches, including time synchronisation, calibration, and preliminary accuracy and precision. Slow- and high-speed activities with different levels of occlusion were evaluated, with performance benchmarked against commercial marker-less MoCap systems, including Qualisys and Theia. A further validation study is currently ongoing, involving 10 healthy participants, to compare the proposed monocular camera-based approach with state-of-the-art marker-based and commercial marker-less motion capture systems, assessing accuracy, precision and participant experience.

The project provided valuable insights into the key technical challenges, appropriate remedial actions, time-synchronisation requirements and the technical capabilities and limitations of the different technologies. These findings have established a strong foundation for larger-scale studies, a robust validation approach plan and future funding applications, including UKRI Translation: MRC Proof of Concept – Stage One, aimed at developing affordable and scalable gait analysis solutions for NHS and primary-care settings, with the potential to improve accessibility, increase patient throughput and reduce waiting times.

Summary of project outcomes

• The project made substantial progress towards its objectives, although some activities were delayed compared with the original plan. BSREC ethical approval required longer than anticipated, and procurement of the sensors also took additional time than expected.
• A key outcome was the generation of a comprehensive training dataset using nine Qualisys cameras, which supported the development and evaluation of both monocular camera-based and mm-Wave motion capture approaches. The pilot experiments provided important technical insights that were not available from our previous work, which had largely focused on static postures. In particular, we established a better understanding of the effective operating range and limitations of mm-Wave sensing during gait and dynamic movement, including the effects of distance. We also systematically explored occlusion effects for both monocular and mm-Wave systems and benchmarked their performance against commercial marker-less solutions.
• The work also identified time synchronisation as an important technical challenge when combining multiple sensing technologies. This has enabled us to identify the underlying issues and develop a practical synchronisation strategy for future experiments.
• The planned validation experiment using the state-of-the-art marker-based system could not be completed within the original project timeframe because of delays in ethical approval and the renovation of the MCIT facility. However, the validation study is now ongoing at the UHCW Gait Laboratory, with the necessary updates incorporated into the BSREC approval.
• Overall, the pilot generated valuable technical knowledge, datasets and an improved validation methodology. These outcomes informed us a larger-scale study and support our planned application to UKRI Translation: MRC Proof of Concept – Stage One, which we plan to submit in November 2026.

Has the project supported further interdisciplinary work

The pilot study generated valuable technical knowledge, datasets and an improved validation methodology, which have informed the design of a larger-scale study and supported our planned application to UKRI Translation: MRC Proof of Concept – Stage One, targeted for submission in November 2026.

The project also established a new collaboration with the UHCW clinical gait team, enabling us to explore the potential application of the technologies in a clinical setting. This collaboration has supported the ongoing validation study at the UHCW Gait Laboratory and provided valuable clinical input into the future development and validation of the proposed systems.

In addition, the pilot study generated new interdisciplinary research ideas around the optimal combination of mm-Wave sensing and monocular camera-based MoCap. Rather than relying on a single sensing modality, combining complementary technologies could improve robustness, accuracy and reliability, particularly under challenging conditions such as occlusion. This provides a basis for future research involving biomechanics, co

Outputs

The pilot study generated valuable technical knowledge, datasets and an improved validation methodology, which have informed the design of a larger-scale study and supported our planned application to UKRI Translation: MRC Proof of Concept – Stage One, targeted for submission in November 2026.

Ongoing generation of a video-based training dataset, which will provide a valuable resource for the computer science community working on ubiquitous and visual computing, computer vision and AI-based human motion analysis.