Generative Adversarial Network

(GAN)

A Seminar Report Submitted

In Partial Fulfilment of the Requirements
for the Degree of

Master of Technology
In
Computer Science
By
Anupam Prakash
(201002)

Under the Supervision of

Dr. Satyabhushan Verma
Department of Computer Science

Babasaheb Bhimrao Ambedkar Universit

(A Central University)
 PREFACE
As a part of the course curriculum and to deepen and widen practical knowledge in the
concept of Generative Adversarial Network (GAN). I was required to make seminar report
on understanding how cryptography work when merged with neural network?

Completing the report helped us know more about the concept Generative Adversarial
Network (GAN). Working with my guide Dr. Satyabhushan Verma I learnt the importance
of cooperation, coordination, and synergy. I hope you will find our seminar report interesting.
All constructive criticism and feedback is cordially invited.

ANUPAM PRAKASH
 ACKNOWLEDGEMENT
In completing this report I have been fortunate to have help, support and encouragement from
many people. I would like to acknowledge them for their cooperation.

First, I would like to thank Dr. Satyabhushan Verma, my project guide, for guiding me
through each and every step of the process with knowledge and support. Thank you for your
advice, guidance and assistance.

Finally, I would like to dedicate this report to my parents for their love, encouragement and
help throughout the project.

ANUPAM PRAKASH

05/07/2021

LUCKNOW
 ABSTRACT
In recent years, supervised learning with convolutional networks (CNNs) has seen huge
adoption in computer vision applications. Comparatively, unsupervised learning with CNNs
has received less attention. In this work we hope to help bridge the gap between the success
of CNNs for supervised learning and unsupervised learning. We introduce a class of CNNs
called deep convolutional generative adversarial networks (DCGANs) that have certain
architectural constraints, and demonstrate that they are a strong candidate for unsupervised
learning. Training on various image datasets, we show convincing evidence that our deep
convolutional adversarial pair learns a hierarchy of representations from object parts to
scenes in both the generator and discriminator. Additionally, we use the learned features for
 Introduction

In recent years deep learning (subset of Artificial Intelligence) gained rapid

success in computer vision. Deep learning technology started in 1980, When
Fukushima started using a convolutional neural network (CNN). Generative
adversarial networks (GANs) are a more recent deep learning development,
invented by Ian Goodfellow and colleagues. GAN is a type of deep learning
model that is aimed at generating new images. GANs are now at the centre of
public attention due to “deepfake” digital media manipulations.

“GAN is designed to create images of objects, animals and human beings based
on what we have in the real world but never existed.”

GANs is an exciting and rapidly changing field, delivering on the promise of

generative models in their ability to generate realistic examples across a range
of problem domains, most notably in image-to-image translation tasks such as
translating photos of summer to winter or day to night, and in generating
photorealistic photos of objects, scenes, and people that even humans cannot tell
are fake.

Generative modelling is an unsupervised learning task in machine learning that

involves automatically discovering and learning the regularities or patterns in
input data in such a way that the model can be used to generate or output new
examples that possibly could have been drawn from the original dataset.

Generative adversarial networks (GANs) are algorithmic techniques which uses

two neural networks, putting one against the other (that is why the word
“adversarial” is used) in order to generate new, synthetic instances of data that
can pass for real data. They are used widely in image generation, video
generation and voice generation. Given a training set, this technique learns to
generate new data with the same statistics as the training set. For example, a
GAN trained on photographs can generate new photographs that look at least
superficially authentic to human observers, having many realistic
characteristics. Though originally proposed as a form of generative model for
unsupervised learning, GANs have also proven useful for semi-supervised
learning, fully supervised learning, and reinforcement learning.

GANs are a clever way of training a generative model by framing the problem
as a supervised learning problem with two sub-models: the generator model that
we train to generate new examples, and the discriminator model that tries to
classify examples as either real or fake. The two models are trained together in a
zero-sum game, adversarial, until the discriminator model is fooled about half
the time, meaning the generator model is generating plausible examples.

GANs’ potential for both good and evil is huge, because they can learn to
mimic any distribution of data. That is, GANs can be taught to create worlds
eerily similar to our own in any domain: images, music, speech, prose. They are
robot artists in a sense, and their output is impressive – poignant even. But they
can also be used to generate fake media content, and are the technology
underpinning Deepfakes.

Artificial
Intelligence
Machine Learning

Deep Learning

Figure1: _____________
Artificial Intelligence- Any Techniques which enables computers to mimic
human beings intelligence by using methods like if- else, if-then else rules and
decision tree. In todays time AI is rapidly growing and used at various fields.

Machine Learning- It is the subset of Artificial Intelligence, It uses

mathematical techniques with previous data or we can called experience to
perform task and improve with time. Models accuracy depends on volume and
quality of data. More the data, more the accuracy.

Deep learning- It is the subset of Machine Learning,

How GANs work-

There are 2 sub models in GANs which is Generator and Discriminator

Generator creates fake data or image in case of image creation by taking

feedback from discriminator.

Discriminator tells how likely the data is similar to real data or not, it acts as a
detective which analyse output and gives feedback. They both learn and
improve together.

Generated

Generator Real of Generated

Discriminator

Real

Figure 2. Schematic model of a generative adversarial network (GAN).

 Related Work
Representation Learning from Unlabled Data

Unsupervised representation learning is a fairly well studied problem in general

computer vision research, as well as in the context of images. A classic
approach to unsupervised representation learning is to do clustering on the data
(for example using K-means), and leverage the clusters for improved
classification scores. In the context of images, one can do hierarchical clustering
of image patches (Adam Coates and Andrew Y. Ng, 2012) to learn powerful
image representations. Another popular method is to train auto-encoders
(convolutionally, stacked (Pascal Vincent and Hugo Larochelle, 2010), these
methods have also been shown to learn good feature representations from image
pixels. Deep belief networks (Honglak Lee and Roger Grosse, 2009) have also
been shown to work well in learning hierarchical representations.

Generating Natural Images

Generative image models are well studied and fall into two categories:
parametric and nonparametric.

The non-parametric models often do matching from a database of existing

images, often matching patches of images, and have been used in texture
synthesis (Alexei A. Efros and Thomas K. Leung, 1999), super-resolution
(William T. Freeman, Thouis R. Jones, and Egon C. Pasztor, 2002) and in-
painting.

Parametric models for generating images has been explored extensively (for
example on MNIST digits or for texture synthesis (Portilla & Simoncelli, 2000).
However, generating natural images of the real world have had not much
success until recently. A variational sampling approach to generating images
(Kingma & Welling, 2013) has had some success, but the samples often suffer
from being blurry. Another approach generates images using an iterative
forward diffusion process (Sohl-Dickstein, 2015). Generative Adversarial
Networks (Goodfellow, 2014) generated images suffering from being noisy and
incomprehensible. A laplacian pyramid extension to this approach (Denton et
al., 2015) showed higher quality images, but they still suffered from the objects
looking wobbly because of noise introduced in chaining multiple models. A
recurrent network approach (Gregor et al., 2015) and a deconvolution network
approach have also recently had some success with generating natural images.
However, they have not leveraged the generators for supervised tasks

Reduce noise in cardiac CTA In multiphase coronary CT angiography (CTA),

a series of CT images are taken at different levels of radiation dose during the
examination. Although this reduces the total radiation dose, the image quality
during the low-dose phases is significantly degraded. Recently, deep neural
network approaches based on supervised learning technique have demonstrated
impressive performance improvement over conventional model-based iterative
methods for low-dose CT. However, matched low- and routine-dose CT image
pairs are difficult to obtain in multiphase CT. To address this problem, we aim
at developing a new deep learning framework.

Method--

We propose an unsupervised learning technique that can remove the noise of the
CT images in the low-dose phases by learning from the CT images in the
routine dose phases. Although a supervised learning approach is not applicable
due to the differences in the underlying heart structure in two phases, the images
are closely related in two phases, so we propose a cycle-consistent adversarial
denoising network to learn the mapping between the low- and high-dose cardiac
phases.
Results- Experimental results showed that the proposed method effectively
reduces the noise in the low-dose CT image while preserving detailed texture
and edge information. Moreover, thanks to the cyclic consistency and identity
loss, the proposed network does not create any artificial features that are not
present in the input images. Visual grading and quality evaluation also confirm
that the proposed method provides significant improvement in diagnostic
quality.

Conclusions-

The proposed network can learn the image distributions from the routine-dose
cardiac phases, which is a big advantage over the existing supervised learning
networks that need exactly matched low- and routine-dose CT images.
Considering the effectiveness and practicability of the proposed method, we
believe that the proposed can be applied for many other CT acquisition
protocols.

Generate brain MR images and test whether radiologists can distinguish

between fake and real

The use of artificial intelligence (AI) has attracted interest for medical imaging
tasks. However, small data sets are major obstacles, in particular for supervised
machine learning and for rare conditions for which only a small number of
cases may exist even in large databases. Even as medical data sets become more
publicly accessible, most of those data sets are restricted to specific medical
conditions, and data collection for machine learning approaches remains
challenging.

To overcome this hurdle, some efforts have turned to the augmentation of

existing data. In this regard, several methods for data augmentation have been
suggested, but minor alterations such as overfitting in learning processes or
geometric transformations have not met the urgent need to provide data sets on
a larger basis. However, considerable progress has been made by the
introduction of synthetic data augmentation to enlarge training sets. By
generating synthetic data, novel created images can be added to existing data
sets. Such an approach may provide a larger number of images to enhance the
variety within a data set and, ultimately, to improve machine learning
algorithms.

Generative adversarial networks (GANs) may contribute toward meeting

the need for synthetic data augmentation. In principle, GANs are based
on an adversarial process where one network is creating artificial images,
while the other network continuously learns to differentiate between real
and generated images. Several studies have applied the concept of GAN
to medical imaging, for example, by producing
New retinal images from a data set of pairs of retinal vessel trees. Apart
from ophthalmologic applications, GANs have also found applicability in
the field of molecular oncology imaging to test the detection rate of
malignant liver lesions by using computed tomography and generated
positron emission tomography (PET) images. However, although current
applications within radiology aim to assist in diagnosis, the quality of
GAN generated data has not yet been validated by human perception in
this context. Thus, in the present feasibility study, we aimed to test the
capability of a GAN to create brain magnetic resonance (MR) images and
to perform quality assessment of GAN generated artificial images by a
visual assessment conducted by blinded radiologists.
Training data of 2-dimensional (2D) images were generated from the
T1W images using the various procedures.