Clear Vision is a desktop application for real-time object detection in video files. It provides a graphical interface that lets users load a video, select a time interval, and run a YOLOv8-based detection pipeline over the chosen frames. Detected objects are highlighted with bounding boxes drawn directly on the video feed.
The application is built on top of the cvision library, which handles the Computer Vision workload, while the Qt 6 GUI manages playback, user interaction, and result visualization.
The project follows a modular structure to separate the core Computer Vision logic from the Graphical User Interface (GUI), ensuring maintainability and scalability.
/gui: Contains the Qt-based implementation./lib: The engine of the application. This is where the integration with cvision happens, wrapping high-level CV tasks.
A Colab notebook with the training process of the YOLOv8 model can be found on YOLO-Treinamento-Ultralytics