Image Stitching using SuperGlue

Fabian Perez · Paula Arguello · Mariana Robayo

Digital Image Processing project at UIS 2024-1

• Visit the report for more information about the project.
• Visit the slides for the presentation.

Overview

This repository contains a Python script for stitching images together using SuperGlue network for matching keypoints and OpenCV for image processing. The script processes a folder of images, aligning and blending them to create a seamless panorama.

Arguments

The stitching.py script accepts the following arguments:

• --folder: Folder path containing images (default: ./acquisitions)
• --crop: Boolean flag to crop the panorama to remove black borders (default: False)
• --blend: Boolean flag to blend the images (default: False)
• --showsteps: Boolean flag to show the steps of the panorama process (default: False)

Image Order

Ensure that the images in the folder are ordered from left to right. It is recommended to name the images sequentially, e.g., image1.jpg, image2.jpg, etc. From the leftmost image to the rightmost image.

Installation

To set up the environment for running the script, follow these steps:

git clone https://github.com/Factral/image-stitching-supeglue
cd image-stitching-supeglue
pip install -r requirements.txt

this code was developed with python 3.10

Running the Script

To create a panorama from the images in the specified folder, use the following command:

python3 stitching.py --folder path/to/your/images --crop --blend --showsteps

you can play with the arguments to see the different results.

Code Structure Explanation

The codebase is structured as follows:

• stitching.py: The main script for stitching images.
• match_pairs.py: A script used for matching keypoints between image pairs using SuperGlue.
• requirements.txt: A file listing the dependencies required to run the script.

License

This project is licensed under the MIT License

feel free to use it and modify it as you wish, if you find this code useful please give a star ⭐ to the repository.