PoseBuddy: workout platform with MoveNet

• Main Author: Cai, Zixin
• Other Authors: Owen Noel Newton Fernando
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2022
• Subjects: Engineering::Computer science and engineering
• Online Access: https://hdl.handle.net/10356/162846
• Institution: Nanyang Technological University

The sedentary lifestyle of modern people is becoming more and more common, and a lack of physical activity is a risk factor for many diseases. Therefore, this project introduces a mobile application that works on iOS and Android, PoseBuddy, through computer vision, provides interactive experiences, and competition through connections with others. It allows users to exercise anytime and anywhere with fragmented time without the limitation of venues. PoseBuddy forms an interactive mode that obtains users' pose data through the mobile device's front camera, inputting to a high-accuracy human pose estimation model provided by TensorFlow, MoveNet. Afterwards, it offers real-time audio feedback during the exercise after being verified by the accuracy validation algorithm. The real-time inputs from the camera feed can be captured and processed asynchronously by the system, allowing users to know in real time if they are carrying out the workout correctly. PoseBuddy is a sports platform enabling users to post their experiences after exercising and making friends.