This project is a fully-featured GUI application for real-time object detection and tracking using YOLOv8. It lets users load videos, preview detections live, customize bounding box colors, and export annotated videos—complete with original audio.
- ✅ YOLOv8 Object Detection and Tracking
- 🖼️ Tkinter GUI with Live Video Preview
- 📂 Drag-and-Drop or File Picker for Input Videos
- 🎨 Customizable Bounding Box Colors (RGB)
- 🔊 Audio Preserved Using FFmpeg (if installed)
- 📊 Progress Bar and Console Logging
- ⚙️ Threaded Video Processing
- 🧠 Automatic CUDA Detection and Usage (if available)
- 📁 Auto-Saved Output in Timestamped Folders
git clone https://github.com/ProjectGlyphMotion/GUI && cd GUIpip install -r requirements.txtEnsure FFmpeg is installed and accessible from the command line.
If you want GPU acceleration, install NVIDIA CUDA and the appropriate drivers.
Cuda can be a pain in the 🍑HOLE if you have a 30 or 40 series card, here is a FIX
python3 GUI.py- Browse or drag a
.mp4,.avi, or.movfile. - Select the desired bounding box color (RGB or via color picker).
- Click ▶ Run Tracker.
- Processed video will be saved in the
output/YYYYMMDD-HHMMSS/directory.
- Format:
<original_filename>_tracked.mp4 - Saved under:
output/YYYYMMDD-HHMMSS/ - If FFmpeg is available, original audio is preserved.
✅ Example Result
MIT License © 2025 Sayan Sarkar & Shitij Halder
Made with love by Sayan and Shitij
This project is based on the Ultralytics YOLOv8, an acclaimed real-time object detection and image segmentation model.