Synthetic Image Dataset Generation for Defense using Generative Adversarial Networks

Yang, Hunmin

doi:2019.22.1.49

J. KIMS Technol > Volume 22(1); 2019 > Article

Research Paper

Journal of the Korea Institute of Military Science and Technology 2019;22(1):49-59.
DOI: https://doi.org/10.9766/KIMST.2019.22.1.049

Synthetic Image Dataset Generation for Defense using Generative Adversarial Networks

Hunmin Yang

Institute of Defense Advanced Research, Agency for Defense Development

국방용 합성이미지 데이터셋 생성을 위한 대립훈련신경망 기술 적용 연구

양훈민

국방과학연구소 국방고등기술원

Abstract

Generative adversarial networks(GANs) have received great attention in the machine learning field for their capacity to model high-dimensional and complex data distribution implicitly and generate new data samples from the model distribution. This paper investigates the model training methodology, architecture, and various applications of generative adversarial networks. Experimental evaluation is also conducted for generating synthetic image dataset for defense using two types of GANs. The first one is for military image generation utilizing the deep convolutional generative adversarial networks(DCGAN). The other is for visible-to-infrared image translation utilizing the cycle-consistent generative adversarial networks(CycleGAN). Each model can yield a great diversity of high-fidelity synthetic images compared to training ones. This result opens up the possibility of using inexpensive synthetic images for training neural networks while avoiding the enormous expense of collecting large amounts of hand-annotated real dataset.

Key Words: Generative Adversarial Networks, Synthetic Image, Deep Learning, Machine Learning, Dataset