Вы находитесь на странице: 1из 7

A Minor Project Synopsis on

Image Restoration
Submitted to Manipal University Jaipur

Towards the partial fulfilment for the Award of the Degree of

BACHELORS OF TECHNOLOGY
In Computers Science and Engineering
2020-2021

By

Manan Uppal
179301112

Under the guidance of

Sandeep Chaurasia

Department of Computer Science and Engineering


School of Computing and Information Technology
Manipal University Jaipur
Jaipur, Rajasthan
Introduction

We’ve all heard the saying A picture is worth a thousand words. But is a tarnished
image with gaping holes or splotches or blurs worth a few hundred? What if you just
found an age-old photograph of your grandparents’ wedding, but the surface was so
worn that you could barely make out their faces. Or perhaps you got photobombed in
what would otherwise have been the perfect picture.

Inpainting is the process of reconstructing lost or deteriorated parts of images and


videos.

Traditional forms of image restoration usually evolve around some simple concept.
Given a gap in pixels, fill the gap with pixels that are the same as, or similar to,
neighbouring pixels. These techniques are generally dependent on various factors and
are more efficient for removing noise or small defects from images. They will most likely
fail when the image has huge gaps or a significant amount of missing data.so I want to
use latest approach using neural network to reduce noise and other defects to bring out
the best picture
Motivation

My grandparents have lots of old memories captured in photos.but even those can’t
withstand the test of time.due to past technology the paint fades away leaving us with a
splotch image where we can’t figure the background or the person.old images of lost
loved ones ,victim of time.Random passerby photobombing what could be a perfect
picture, all these problems exist and with the help of neural networks I want to tackle
these problems
Objective

1) Model-based optimization methods are usually time-consuming with sophisticated


priors for the purpose of good performance.

2) While discriminative learning methods have fast testing speed but their application
range is greatly restricted by the specialized task.

3) This project aims to train a set of fast and effective CNN (convolutional neural
network)
denoisers and integrate them into model-based optimization method to solve other
inverse problems. My objective is to try and experiment so that the learned set of
denoisers can not only achieve promising Gaussian denoising results but also can be
used as prior to deliver good performance for various low-level vision applications.
Methodology/planning of work

My plan of work will be a methodological approach different steps will be


1) Research about various neural networks
2) Researching about CNN’s and how will they be applied as denoisers
3) Creating a simple prototype networks
4) Creating a first working prototype
5) Increasing the accuracy of the developed network
6) Final working prototype/product
Facilities required for proposed work:

For every project there are requirements which are vital for a project to be completed,
my project is no different. I would be obliged to have the under-mentioned facilities
made available to me by the university.

Facilities required are:


1)Software
1.1)python3
1.2)Tensorflow API,pytorch API
1.3)Open cv
1.4)MATLAB
1.5)Cuda & cuDNN
1.6)MatConvNet

2)Hardware
2.1)Cloud instance(GPU)
2.2)Local Machine
Bibliography/References

1) http://openaccess.thecvf.com/content_cvpr_2017/papers/
Zhang_Learning_Deep_CNN_CVPR_2017_paper.pdf

2) https://ieeexplore.ieee.org/abstract/document/5597860

3) https://link.springer.com/article/10.1007/s10851-007-0017-6

4)https://heartbeat.fritz.ai/guide-to-image-inpainting-using-machine-learning-to-edit-and-
correct-defects-in-photos-3c1b0e13bbd0

5) https://www.google.com/url?
sa=t&rct=j&q=&esrc=s&source=web&cd=3&cad=rja&uact=8&ved=2ahUKEwjYjbzsu5jn
AhVWzTgGHa8lBEYQFjACegQIAhAB&url=https%3A%2F%2Ftowardsdatascience.com
%2Fimage-inpainting-humans-vs-
ai-48fc4bca7ecc&usg=AOvVaw1UdlL3OAYJa7Nu9TVVwnby

6)https://www.google.com/url?
sa=t&rct=j&q=&esrc=s&source=web&cd=4&cad=rja&uact=8&ved=2ahUKEwjYjbzsu5jn
AhVWzTgGHa8lBEYQFjADegQICBAB&url=https%3A%2F%2Fwww.nvidia.com%2Fres
earch%2Finpainting%2F&usg=AOvVaw3s_QhUCuGUsvApjie8RiIr

7) https://www.google.com/url?
sa=t&rct=j&q=&esrc=s&source=web&cd=7&cad=rja&uact=8&ved=2ahUKEwjYjbzsu5jn
AhVWzTgGHa8lBEYQFjAGegQIBxAB&url=https%3A%2F%2Fmedium.com%2Fjamieai
%2Fimage-inpainting-with-deep-learning-
dd8555e56a32&usg=AOvVaw2Z5fqQn0qdFmx0OS3ZD0aK

Вам также может понравиться