report

Acknowledgements	
Abstract	
1.	Introduction	
1.1 Artificial Intelligence	
1.2 Artificial Neural Network	
1.2 Machine Learning	
1.4 Animation Character Design	
2.	Literature Review	
2.1 GAN Model	
3.	System Design	
3.1 GAN design	
4.	Methodology and resources	
4.1 data collection	
4.2 training	
5.	Implementation	
5.1 Using PyTorch	
6.	Results	
6.1 GAN outputs	
7.	Conclusion	
Cited Reference	

 
1. Introduction

The recent years has witnessed a boom in the development of artificial intelligence 
and its applications.  Especially the method of neutral network is getting increasing 
attention these days. though first proposed in 1940s, it was not until the advancement of 
parallel computing and GPU that it became so important. GAN, short for Generative Adversarial
network, is perhaps the most renowned and powerful tool in this realm. It has been proved
effective in generating photos of people, cats...even dialogs... In this paper we try 
to use GAN to generate animation character designs.


1.1 Artificial Intelligence	

Artificial intelligence (AI), or machine intelligence, is intelligence demonstrated by machines, in contrast with natural intelligence, which is displayed by humans.  First founded as an academic discipline in 1955, artificial intelligence has waxed and waned, experiencing several waves of optimism, followed by disappointment and the loss of funding, and then by new approaches, success and renewed funding.In the twenty-first century, thanks to concurrent advances in computer power, large amounts of data, and theoretical understanding , AI techniques met its resurgence again and thus have become an essential part of the technology industry, showing its power in solving many challenging problems in computer science, software engineering and operations research.

1.2 Artificial Neural Network	

Artificial neural networks (ANNs, usually simply “neural networks” or NNs), are computing systems mimicking the biological neural networks that constitute animal brains. An ANN consists of a collection of connected units or nodes (called artificial neurons), which loosely model the neurons in a biological brain. Each connection, called edges, which resembles the synapses in a biological brain, transmits signals to other neurons. An artificial neuron that receives a signal then processes it and can signal neurons connected to it. The "signal" at a connection is a real number, and the output of each neuron is computed by some non-linear function of the sum of its inputs. Neurons and edges typically have a weight that can adjust while learning. The weight increases or decreases the strength of the signal at a connection. Neurons may possess a threshold value that limits the sending of a signal: a signal is sent only if the aggregate signal crosses that threshold. Typically, neurons are aggregated into layers. Different layers may perform different transformations on their inputs. Signals travel from the first layer (the input layer), to the last layer (the output layer), possibly after traversing the layers multiple times.

Famous models includes ... GAN ...

1.3 Machine Learning	

Machine learning (ML), as a subset of artificial intelligence, deals with computer algorithms that improve and adjust itself automatically through experience.  It adopts a mathematical model based on sample data(as "training data") to make predictions or decisions rather than  being intentionally devised and programmed to do so. Machine learning algorithms are very suitable in situations where it is difficult or infeasible to develop conventional algorithms to perform the needed tasks.

1.4 Animation Character Design