Inverse Imaging: Reconstructing High-Resolution Images from Degraded Images

EasyChair Preprint 7847

Abstract

Image Denoising simply follows the U-Net architecture which uses images with noise. In this work, we show that u-net architecture, based on convolutional and deconvolutional or transpose convolutional neural networks does a pretty good job in removing noise from the image. The task belongs to a general class of problems on Posterior probability distribution that is the probability of the parameter theta given the evidence X: P(theta | X).

Keyphrases: Convolutional Neural Networks, Deconvolutional Neural Networks, Image denoising, U-Net architecture, deep learning

@booklet{EasyChair:7847,
  author    = {Samir Mishra and Mrunal Chide and Aishwarya Manmode and Rohit Pandita and Mansi Bhonsle},
  title     = {Inverse Imaging: Reconstructing High-Resolution Images from Degraded Images},
  howpublished = {EasyChair Preprint 7847},
  year      = {EasyChair, 2022}}