Image Super Resolution (2nd Winner Daisi Hackathon)

Python function as a web service to enhance low resolution image to high quality image.

The service is based on Single Image Super Resolution (SISR) deep learning model which is available on Open Model Zoo, check this link for more info.

Super Resolution is the process of enhancing the quality of an image by increasing the pixel count using deep learning.

• The model (Neural Network) expects inputs with a width of 480, height of 270.
• The model returns images with a width of 1920, height of 1080.
• The image sides are upsampled by a factor 4. The new image is 16 times as large as the original image.

In brief:

• Input image should be: 480x270 resolution.
• Output image : 1920x1080 resolution.

You can use image samples in the /images directory to test it on the model.

How to call it

-Load the Daisi

import matplotlib.pyplot as plt
import pydaisi as pyd
image_super_resolution = pyd.Daisi("oghli/Image Super Resolution")

-Call the image_super_resolution end point, passing the image source to enhance it, you can pass image source either from images/ directory or from valid url of the image

#image_source = "https://i.imgur.com/R5ovXDO.jpg"
image_source = "images/witcher.jpg"
result = image_super_resolution.cv_superresolution(image_source).value
result

-It will return two np arrays representing:

• Original image reshaped to the target resolution of the model
• Super resolution image

you can save result images in variables

origin_image = result[0] 
superresolution_image = result[1] 

Then display super resolution image

plt.imshow(superresolution_image)

Also, you can display comparison figure for image before and after enhancement by excuting the following code

fig, ax = plt.subplots(nrows=1, ncols=2, figsize=(30, 15))
ax[0].imshow(origin_image)
ax[1].imshow(superresolution_image)
ax[0].set_title("Origin")
ax[1].set_title("Superresolution")

Review this notebook for the full demo.

Function st_ui included in the app to render the user interface of the application endpoints.

References

Check this research paper reference for more info about super resolution model.