Abstract: Weight training and physical therapy depend strongly on correct form in order to achieve better results. Even more important, incorrect execution can lead to undesired results such as slow muscle growth or serious injury. Taking advantage of the limb-detection capabilities of the state of the art computer vision algorithms, we can prevent these negative effects and ensure optimal training. This project proposes to create a self-contained assistive application that will observe the user while they are training, analyze their skeletomuscular structure through computer vision, and offer form assistance and personalized advice to increase performance while avoiding injury.
The future of human-software interaction will be much more seamless and context-aware in the future. Particularly, embodied assistant intelligence software, taking the forms of androids and drones, will be more prevalent. Therefore, we will demonstrate the value of such a future by deploying our physical training assistance software both by itself on a smartphone and also coupled with small drone.
Log in or sign up for Devpost to join the conversation.