This thesis presents the results of a practical internship experience conducted at BBSoF, a startup affiliated with the University of Padua, aimed at developing a mobile application to analyze squat exercises and gait patterns. The mobile app leverages Flutter as the development framework and integrates machine learning capabilities using Google's Mediapipe library. The primary objective of the application is to provide users with real-time feedback and valuable insights into their exercise performance by identifying both correct and incorrect movement patterns. This methodology involves the fusion of computer vision techniques with biomechanical analysis to accurately evaluate user movements. By processing video recordings of squat exercises and gait patterns, the application extracts key metrics related to posture, joint angles, and alignment. These metrics are then analyzed to generate personalized feedback, highlighting areas of improvement and reinforcing correct techniques. The results of this thesis demonstrate the effectiveness of combining machine learning models and computer vision with biomechanical analysis of movements to improve the performance of various movements and prevent potential injuries.
Intelligenza Artificiale per il Fitness: sviluppo di una Mobile App per l'Analisi del Movimento tramite Computer Vision
NEFFAT, DAVIDE
2023/2024
Abstract
This thesis presents the results of a practical internship experience conducted at BBSoF, a startup affiliated with the University of Padua, aimed at developing a mobile application to analyze squat exercises and gait patterns. The mobile app leverages Flutter as the development framework and integrates machine learning capabilities using Google's Mediapipe library. The primary objective of the application is to provide users with real-time feedback and valuable insights into their exercise performance by identifying both correct and incorrect movement patterns. This methodology involves the fusion of computer vision techniques with biomechanical analysis to accurately evaluate user movements. By processing video recordings of squat exercises and gait patterns, the application extracts key metrics related to posture, joint angles, and alignment. These metrics are then analyzed to generate personalized feedback, highlighting areas of improvement and reinforcing correct techniques. The results of this thesis demonstrate the effectiveness of combining machine learning models and computer vision with biomechanical analysis of movements to improve the performance of various movements and prevent potential injuries.File | Dimensione | Formato | |
---|---|---|---|
Neffat_Davide.pdf
embargo fino al 09/07/2027
Dimensione
4.34 MB
Formato
Adobe PDF
|
4.34 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/66517