The project, VolleyVisionAI, is a web app designed for video analysis in the sport of volleyball. Its main objective is to bridge the market gap between professional tools, which are often inaccessible, and the need for an intuitive and user-friendly application aimed at amateur users.
VolleyVisionAI operates in two modes: a manual mode, which enables users to create, import, and export video analysis projects, perform event tagging, and generate customized video clips. Additionally, users can draw directly on the video player thanks to a simple dashboard, enhancing the precision of their analysis. The second mode, AI mode, leverages Computer Vision models to track the ball and recognize players on the court. The project has been built on an architecture that ensures high performance and scalability, utilizing React.js for the front-end and FastAPI for the back-end.
The development process involved a thorough requirements analysis, the creation of an initial prototype using Figma, and the technical development of both the user interface in JavaScript and the back-end functionalities in Python. The project also integrates AI-driven analysis through the YOLOv9 recognition model and the OpenCV library.
Usability testing has validated the effectiveness of the application and the intuitiveness of its interface. Feedback received during the testing phase has further improved the user experience and provided insights for future development. The beta version of VolleyVisionAI is fully operational and has successfully achieved its objectives. The outcomes open new opportunities for future enhancements, including the addition of new AI features and expanding accessibility through a robust deployment system.