We present a solution for intelligent posture training based on accurate, real-time sitting posture monitoring using the LifeChair IoT cushion and supervised machine learning from pressure sensing and user body data. We demonstrate our system’s performance in sitting posture and seated stretch recognition tasks with over 98.82% accuracy in recognizing 15 different sitting postures and 97.94% in recognizing six seated stretches. We also show that user BMI divergence significantly affects posture recognition accuracy using machine learning. We validate our method’s performance in five different real-world workplace environments and discuss training strategies for the machine learning models. Finally, we propose the first smart posture data-driven stretch recommendation system in alignment with physiotherapy standards.

We propose a system for self-supported martial arts training using an IoT sensing platform and Serious Game that can also be extended for general sports training. In martial arts, it is important that the practitioner is correctly performing each technique to accurately learn and prevent injury. A common stance in all martial arts, but especially in Shaolin Kung Fu, is the horse stance or Mabu. With the pandemic, many more people adopted remote training without the presence of a professional trainer to give advice. Our developed LifeMat system, which is a novel IoT pressure-sensitive training mat, uses pressure maps and pattern recognition to accurately classify key martial arts postures, provide feedback on form, and detect when the user performs the technique incorrectly. This is presented in the form of a Serious Game we have developed named Kung Future that focuses on the Mabu stance as a case study. We tested 14 participants with three different feedback conditions and demonstrated that, on average, participants had higher performance, duration, engagement, and motivation when game feedback was active. Furthermore, user responses from the surveys suggested positive feedback for real-world and long-term use and scalability.

We propose a platform for independent remote training in Martial Arts (MA) using an IoT sensing mat, virtual AI coach and Serious Game that can be extended for sports training. In MA, it is important to accurately train a technique to prevent injury. Our platform uses pressure maps and pattern recognition to accurately classify MA postures and provide real-time feedback. We also present our Serious Game named Kung Future that includes an AI coach and focuses on training the Mabu stance of Kung Fu as a case study. Results from user studies with 14 participants with different feedback conditions demonstrated that participants had improved technique performance when training with our platform. Responses from the survey showed positive adoption for real-world and long-term use. We presented the platform at the Tokyo Dome in a public showcase and validated the AI coaching interface through user testing.

This paper presents LifeChair-X, an IoT cushion for posture-based gaming and Serious Game for improving workplace wellness in a post-pandemic world of sedentarianism. The LifeChair posture models are used as input to a game in which a user moves an avatar using postural shifts. Through consultations with medical professionals at the University of Tsukuba Hospital, we developed the game for guiding seated exercises for core training. The system is evaluated with 12 adult participants. Experiments showed the LifeChair-X average response time was within 0.15 s (SD=0.21) of a conventional gamepad. A gameplay study demonstrated an average score of 53.50% using LifeChair-X which was less than 25% difference to a gamepad. Elevated heart rates were observed across all participants when using LifeChair-X suggesting light-intensity exercise. A system usability score of 88.96 (SD=10.12) was achieved suggesting excellent usability levels. The LifeChair-X is an effective system for exergaming and can improve workplace wellness.

Physical inactivity is still a major problem contributing to a growing public health crisis despite a fast-expanding body of technological solutions and wellness research around fitness training. The inaccessibility of professional fitness training remains a leading cause of this gap for reasons encompassing socioeconomic factors, cultural and demographic barriers, and more recently the threat of global pandemics that disrupt traditional modes of staying physically active. Previous lines of work have explored using AI for fitness activity recognition from various sensing modalities such as computer vision, wearable sensors, and force and pressure sensors. However, these works are limited by their feasibility, deployability, and accessibility in real-world scenarios, in addition to the technical challenges faced by each modality for accurate and reliable activity recognition. In this paper, we propose an accessible system for gym activity recognition and correction focusing on foundational fitness activities using ML and a novel pressure sensing mat, and validate its deployability in a real-world use case in a natural gym setting. We present the detailed and previously under-investigated Centre of Pressure (COP) profile of four main gym activities in terms of several COP-related metrics specifically as targets for ML-based recognition tasks. Based on this, we identify COP displacement and COP balance measures as important features for ML-based recognition of these fitness activities for future research in this area. Furthermore, we compare the performance of several ML models in the activity recognition task, achieving 98.5% recognition accuracy using ML models suitable for real-time deployment. Finally, we demonstrate the feasibility of our system in a live real-world with use case in a natural gym environment.