Image stitching is the process of combining multiple images with overlapping fields of view to produce a segmented panorama. Here, we provide
ImageStitcher to easily stitch a number of images.
Learn more: https://paperswithcode.com/task/image-stitching/
from kornia.contrib import ImageStitcher matcher = KF.LoFTR(pretrained='outdoor') IS = ImageStitcher(matcher, estimator='ransac').cuda() # NOTE: it would require a large CPU memory if many images. with torch.no_grad(): out = IS(*imgs)
Explore with your data: https://colab.research.google.com/github/kornia/tutorials/blob/master/source/image_stitching.ipynb
Visit the demo on Hugging Face Spaces.