Project GlyphMotion 🌀✨

This script performs object tracking on input videos using YOLOv8 and allows you to specify bounding box colors. It supports threading for I/O, system resource monitoring, and preserves original video audio using FFmpeg.

---

🚀 Features

• ✅ YOLOv8 Object Detection and Tracking
• 🎨 Customizable Bounding Box Color
• ⚙️ Threaded Frame Reading/Writing
• 📊 Real-Time CPU/GPU/RAM Monitoring
• 🧠 GPU Acceleration (CUDA supported)
• 🔈 Preserve Audio using FFmpeg
• 💾 Automatic Output Directory Creation
• 🔄 Progress Bar with tqdm

---

🛠️ Setup

Clone the Repository

git clone https://github.com/ProjectGlyphMotion/Script && cd Script

Install Dependencies

pip install -r requirements.txt

• Ensure FFmpeg is installed and accessible from the command line.

• If you want to use GPU power to process the video, ensure NVIDIA CUDA is installed properly.

• Cuda can be a pain in the 🍑HOLE if you have a 30 or 40 series card, here is a FIX

---

🧪 Usage

Run the script and enter the input video path when prompted:

python3 ot.py                      

For custom RGB colours (Default is Green)

🔵 Blue

python3 ot.py --box_color 0,0,255  

🔴 Red

python3 ot.py --box_color 255,0,0  

🟢 Green

python3 ot.py --box_color 0,255,0  

📂 Output

• Videos are saved in the output/ directory.
• Format: <original_name>_processed.mp4

✅ Example Result

---

📃 License

MIT License © 2025 Sayan Sarkar & Shitij Halder

Credits

Made with ❤️ by Sayan and Shitij

This project is based on the Ultralytics YOLOv8, an acclaimed real-time object detection and image segmentation model.