Real Time Object Detection and Communication

This project implements real-time object detection and communication using WebRTC. The system consists of a client and a server that communicate with each other to perform object detection on video frames captured from a webcam. The client then asynchronously processes the video frames using a pre-trained YOLOv5 model and displays the video of the frames with and without the bounding boxes.

Project Structure

The project is organized as follows:

• client.py: Client-side code for real-time object detection and WebRTC communication.

• server.py: Server-side code for processing detected coordinates from the client.

• rtc_signal_handlers.py: Signaling handlers for WebRTC communication.

• requirements.txt: Project dependencies.

• Dockerfile_client: Dockerfile for the client-side image.

• Dockerfile_server: Dockerfile for the server-side image.

Run using python

python server.py --signaling tcp-socket

And in another terminal on the same machine:

python client.py --signaling tcp-socket

Building Docker Images

To build the Docker images for both server and client, run the following commands in the project directory:

docker build -f Dockerfile_server -t webcam_server .
docker build -f Dockerfile_client -t webcam_client .