Segmentation İnferences Tool for U-Net and YOLOv7

This project was conducted to compare the two most widely used segmentation architectures.

The system features:

OS : Ubuntu 20.04 LTS 64-bit 
CPU : Intel(R) Core(TM) i5-10200H CPU @ 2.40GHz
GPU : Nvidia GTX 1650ti 4GB
RAM : Samsung M471A1K43DB1-CWE 16GB

Inferences on Photo and Video

U-Net

YOLOv7

Cloning the Repository

Clone this repository with git.

  git clone https://github.com/oguzaybilir/Lane-Segmentation.git
  cd Lane-Segmentation

Installing Required Libraries

There is a requirements.txt file to install packages you need. This file contains almost all libraries and modules used in the project.

To install this libraries and packages:

    pip3 install -r requirements.txt

Run

The tool is so easy to use.

  python3 main.py --weights "path to your YOLOv7 weights" --source "path to your photo or video"
  python3 main.py --weights "path to your U-Net weights" --source "path to your photo or video"

This repository only accepts source files as .mp4, .jpg, .png and only accepts weight files as .pt, .h5 and .hdf5 . But you can adjust the extensions in the main.py and segment/predict.py

Authors

• @oguzaybilir

Special Thanks and Regards

I manipulated Rizwan Munawar's https://github.com/RizwanMunawar/yolov7-segmentation repository a little bit for my usage. So I owe a thank you to him.

I also owe a debt of gratitude to my mentor Mehmet Okuyar. He guided me in my journey in the fields of Artificial Intelligence and Image Processing.

Acknowledgements

• Mehmet Okuyar
• RizwanMunawar