Accurate pose estimation and exercise recognition for next generation wearable systems for athletic training

点击翻译按钮获取中文摘要

In the last 5 years, there has been a rapid expansion in the use of wearable systems for exercise and activity monitoring, including such popular systems as the FitBit and Garmin vivo. These systems enable the user to track their exercise activity, log and monitor their activity and performance over time, as well as engage in social games such as virtual team workouts and challenges. However, the majority of these systems can only measure a very rough estimate of overall activity, and are unable to provide more detailed measurement and assessment of specific workout activities to provide targeted feedback. In this proposal, our objective is to develop and validate the next generation wearable system for detailed and accurate exercise activity tracking. The developed system will accurately measure the body position of the athlete as they are performing their training exercises, and provide detailed feedback to the athlete and their coach. This will allow athletes and coaches to carefully track and analyse the athlete's form, performance and progress. Such detailed tracking and immediate feedback will help to reduce and prevent injury, improve performance and improve the quality of training. The project is a collaboration between PUSH Inc., a Canadian start-up based in Toronto, Ontario, focused on the development of wearable fitness devices for athletic training, and researchers at the University of Waterloo. The developed system and technology will be transferred to PUSH to enable them to bring to market the next generation of wearable systems for athletic activity measurement and monitoring.