AI monitors movements to aid rehab
Researchers have developed an AI-powered system to assess rehabilitation exercises in real time, aiming to make computer-assisted therapy more practical on devices with limited processing capabilities. The system is called RMPE Tiny; it is a human pose estimation network that uses computer vision techniques to measure body movements, including key points such as shoulders, knees, and ankles. This system was designed to reduce computational requirements without significantly compromising accuracy and has achieved an accuracy of over 96% in pose estimation. It is potentially suitable for future use in supporting rehabilitation systems that provide immediate feedback outside clinical settings.
Size: 120%Text Size
Notice: This Is an AI-Generated Summary
Comments (0)