From movement analysis and exercise recognition to real-time coaching and mobile deployment, we build AI systems for connected fitness products, training platforms, and professional performance environments.
Computer vision systems that track body position, joint movement, posture, and exercise technique using standard or specialised cameras.
AI models that recognise exercises, count repetitions, measure movement patterns, and identify whether activities are performed correctly.
Immediate visual feedback that helps users improve technique, maintain proper posture, and reduce potentially unsafe movement.
Multimodal analysis of movement, attention, effort, engagement, and other behavioural signals during training.
Efficient AI models deployed on smartphones, connected fitness equipment, wearables, and other edge devices for responsive, privacy-conscious analysis.
Tailored AI solutions built around specific sports and fitness applications, from initial feasibility studies to production deployment.
Examples of how machine learning and computer vision can help transform sports events, sport tech and help pave the way for better results and experiences.
Training platforms often lack insight into how users are actually performing and responding
Digital training, coaching, and simulation platforms can track basic metrics such as repetitions, completion rates, or session duration, but often have limited understanding of movement quality, engagement, cognitive workload, stress, or frustration. Without this context, it is difficult to adapt training effectively to the individual user or identify when performance is being affected by physical or cognitive factors.
Multimodal AI for real-time performance and behavioural analysis
Computer vision and machine learning can combine camera, microphone, and wearable data to analyse movement together with behavioural and cognitive-affective signals such as attention, engagement, workload, stress, and frustration.
Integrated into a training platform, these insights can be used to dynamically adjust content, difficulty, pacing, or feedback according to the user’s current performance and state. The technology can support digital coaching applications, fitness platforms, professional simulators, and other connected training systems while being developed as reusable AI components for integration into existing products.