retinal-research / ote-gan

Code for the paper "OPTIMAL TRANSPORT GUIDED UNSUPERVISED LEARNING FOR ENHANCING LOW-QUALITY RETINAL IMAGES"

OTE-GAN

Code for the paper [ISBI '2023] Optimal Transport Guided Unsupervised Learning for Enhancin Low-Quality Retinal Images

Here is latest work for OT retinal denoising

[IPMI '2023] OTRE: Where Optimal Transport Guided Unpaired Image-to-Image Translation Meets Regularization by Enhancing) With Github

Visual Results

License

Released under the ASU GitHub Project License.

ote-gan's Issues

数据集

您好，请问数据集怎么导入？

How can I use the Model

is there a link to the pre-trained weights and how I can use the repository

No-Reference Quality Assessment.

Question 1:

I'm trying to understand the CR metric. Does it represent the percentage of all enhanced images that are considered high-quality (compared to both high-quality and usable results)?

Question 2:

The other metrics mentioned (Accuracy, Kappa, AUC) seem to relate to a model trained for classifying images as high-quality, usable, or rejected only and there is no relation in enhancement here am I missed something?

full reference experiment

I want to ask about rows for example the row "ours" this calculated between the degraded images and their corresponding enhancement images or (between the original high quality images and their enhancement)