<<<<<<< HEAD The aim of this project is try reproduce a research paper explores the possibilities of context-aware applications within the strength training domain. It does so by analyzing wristband accelerometer and gyroscope data obtained during strength training sessions. The collected dataset contains data of 5 participants performing various barbell exercises. The goal is to explore, build, and evaluate models that can, just like human personal trainers, track exercises, count repetitions, and detect improper form. The methods evaluated in this paper use a supervised learning approach for classification. Various machine learning algorithms were trained using the collected dataset and accuracies were compared to find and evaluate the right models.
During the last decade, many practical constraints related to carry-on sensors like accelerometers, gyroscopes and GPS-receivers have been solved. This has enabled monitoring and classification of human activity based on information from wearables such as smartwatches to become a growing research area of pattern recognition and machine learning. The grounds for that lie in high commercial potential of context-aware applications and user interfaces. Moreover, activity recognition can be utilized to attack some of the serious societal challenges like rehabilitation, sustainability, elderly care and health. To promote healthier lifestyles, works from the past have focused on tracking movements and user feedback via exercise management systems. Such systems partly replace tasks that are currently fulfilled by personal trainers. For example,
in the category of aerobic exercises such as bicycling, swimming, and running, there are accelerometer and GPS-based pedometers to track running pace and distance, ECG monitors to track exertion, and electronic exercise machines such as treadmills, elliptical trainers, stair climbers, and stationary bikes. Weight training, in addition to aerobic exercises, is an important component of a balanced exercise program.
8ab2d12b32bd5206b140bc6dd7cee5bc2ce0d497