Readmore_registration.md

Fundus image registration

Match/Align a fundus photograph to another fundus photograph from the same eye.

See the usage_registration.ipynb.

[x] Works on tensor images
[ ] Has batch support

Example: Alignment of the moving image to the fixed image

Ref

This is a wrapper for SuperRetina from Jiazhen Liu et al., "Semi-Supervised Keypoint Detector and Descriptor for Retinal Image Matching" (2022). This package provides an interface to the registration task. Weights will be aquired automatically from zenodo. The input image contrasts are enhanced before being fed into the SuperRetina model at a size of 512x512 pixels.

Cite

@InProceedings{superretina_liu_2022,
    author="Liu, Jiazhen
    and Li, Xirong
    and Wei, Qijie
    and Xu, Jie
    and Ding, Dayong",
    title="Semi-supervised Keypoint Detector and Descriptor for Retinal Image Matching",
    booktitle="Computer Vision -- ECCV 2022",
    year="2022",
    publisher="Springer Nature Switzerland",
    pages="593--609",
    isbn="978-3-031-19803-8"
}