Deceptrax123 / autoencoder-image-glare-removal

Autoencoder that removes glare on images by applying an Inpaint effect to the image

Repository from Github https://github.comDeceptrax123/autoencoder-image-glare-removalRepository from Github https://github.comDeceptrax123/autoencoder-image-glare-removal

Deep Learning Based Glare Removal from Images

This repository is a part of Spurious Correlation Mitigation. It contains the code for a base autoencoder model used to learn the distribution of images without sunlight and generate the same given an input with many glare spots.

Two Models were trained:

  1. Model to reduce brightness to a specific level
  2. Model to reduce brightness through image inpainting.

The models have been trained and implemented using PyTorch.

About

Autoencoder that removes glare on images by applying an Inpaint effect to the image


Languages

Language:Jupyter Notebook 98.5%Language:Python 1.5%