NYC Traffic Detector by Cole Feuer

About the Project

Cole Feuer developed the NYC Traffic Detector to automatically identify and classify traffic infrastructure—such as stop signs, traffic lights, and vehicles—in real‑time video feeds. Leveraging a YOLOv5 model fine‑tuned on NYC traffic datasets and OpenCV for video processing, this tool aims to support smart city analytics and autonomous navigation research.

Key Features

• Real‑time inference at 30+ FPS on commodity GPUs
• Detects and classifies traffic signs, signals, and vehicle types
• Built with PyTorch, YOLOv5, and OpenCV for efficient video handling
• Modular pipeline: easy to retrain on custom datasets
• Dockerized for seamless deployment on edge devices

Installation & Usage

1. Clone the repository:
   git clone https://github.com/CDFire/TrafficDetector.git
2. Install dependencies:
   pip install -r requirements.txt
3. Download pre-trained weights:
   bash scripts/download_weights.sh
4. Run detection on video:
   python detect.py --source path/to/video.mp4

Contact & Connect

For questions, feedback, or collaboration, reach out to coledfeuer@gmail.com or connect on LinkedIn.