Image-Stitching-with-Keypoint-Descriptors

It is aimed to obtain a final panorama image that will combine the subimages provided using key point identification methods (SIFT/SURF and ORB) and include all the scenes in the subimages.

IMPORTANT INFORMATION

Since the method I used to create a panorama image from subimages is a bit long, you can export the dataset from a small number of images.

Requirement

• python3 (or higher)
• opencv 3 (or higher)

You will need to install some package using pip:

• numpy
• matplotlib

Usage

$ python main.py <feature extraction method name(SIFT or ORB)>

for example

$ python main.py SIFT $ python main.py ORB $ python main.py // SIFT is selected manually

Input format

The dataset folder must be in the same directory as main.py

Output

The program will output:

• Showing every feature extraction, matched and Registration images by using matplotlib. If you don't want to see it. Comment lines 45, 50, 67, 94 in merge.py.
• A panoma stitched images Stitched_Panorama_Full_{pano_name}.png

Environment

I test my code in Window11. It should work fine in these system.