---
license: apache-2.0
tags:
- image-segmentation
- vision
- fundus
- optic disc
- optic cup
widget:
- src: >-
    https://huggingface.co/pamixsun/swinv2_tiny_for_glaucoma_classification/resolve/main/example.jpg
  example_title: fundus image
---
# Model Card for Model ID

<!-- Provide a quick summary of what the model is/does. -->

This model utilizes a Swin Transformer architecture and has undergone supervised fine-tuning on retinal fundus images from the [REFUGE challenge dataset](https://refuge.grand-challenge.org/). 
It is specialized in automated analysis of retinal fundus photographs for glaucoma detection. 
By extracting hierarchical visual features from input fundus images in a cross-scale manner, the model is able to effectively categorize each image as either glaucoma or non-glaucoma. Extensive experiments demonstrate that this model architecture achieves state-of-the-art performance on the REFUGE benchmark for fundus image-based glaucoma classification. 
To obtain optimal predictions, it is recommended to provide this model with standardized retinal fundus photographs captured using typical fundus imaging protocols.

## Model Details

### Model Description

<!-- Provide a longer summary of what this model is. -->

- **Developed by:** [Xu Sun](https://pamixsun.github.io)
- **Shared by:** [Xu Sun](https://pamixsun.github.io)
- **Model type:** Image classification
- **License:** Apache-2.0

## Uses

<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->

his pretrained model enables semantic segmentation of key anatomical structures， namely, the optic disc and optic cup, in retinal fundus images. 
It takes fundus images as input and outputs the segmentation results.

## Bias, Risks, and Limitations

<!-- This section is meant to convey both technical and sociotechnical limitations. -->

The model has undergone specialized training and fine-tuning exclusively using retinal fundus images, 
with the objective to perform semantic segmentation of anatomical structures including the optic disc and optic cup. 
Therefore, in order to derive optimal segmentation performance, it is imperative to ensure that only fundus images are entered as inputs to this model.

## How to Get Started with the Model

Use the code below to get started with the model.

```python
import cv2
import torch

from transformers import AutoImageProcessor, SegformerForSemanticSegmentation

image = cv2.imread('./example.jpg')
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

processor = AutoImageProcessor.from_pretrained("pamixsun/segformer_for_optic_disc_cup_segmentation")
model = SegformerForSemanticSegmentation.from_pretrained("pamixsun/segformer_for_optic_disc_cup_segmentation")

inputs = processor(image, return_tensors="pt")

with torch.no_grad():
    outputs = self.seg_model(**inputs)
    logits = outputs.logits.cpu()

upsampled_logits = nn.functional.interpolate(
    logits,
    size=image.shape[:2],
    mode="bilinear",
    align_corners=False,
)

pred_disc_cup = upsampled_logits.argmax(dim=1)[0].numpy().astype(np.uint8)

```

## Citation [optional]

<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->

**BibTeX:**

[More Information Needed]

**APA:**

[More Information Needed]


## Model Card Contact

- pamixsun@gmail.com