期刊论文、会议论文

Yu, Xinguo

英文

Lecture Notes in Computer Science

0302-9743

2019

11854

211-224

9th Pacific-Rim Symposium on Image and Video Technology, PSIVT 2019

Image and Video Technology

18 November 2019 through 22 November 2019

Sydney, AUSTRALIA

[Ullah, Anwar;Yu, Xinguo;Majid, Abdul] Cent China Normal Univ, Natl Engn Res Ctr E Learning, Wuhan, Peoples R China.^[Rahman, Hafiz Ur] Guangzhou Univ, Sch Comp Sci, Guangzhou 510006, Peoples R China.^[Mughal, M. Farhan] Tianjin Univ Finance & Econ, Tianjin, Peoples R China.

Chilwoo Lee<&wdkj&>Zhixun Su<&wdkj&>Akihiro Sugimoto

GEWERBESTRASSE 11, CHAM, CH-6330, SWITZERLAND

Springer

9783030348786

10.1007/978-3-030-34879-3_17

Keywords: Generative Adversarial Network (GAN) · Iterative GAN · Text-to-image synthesis · CUB dataset · Oxford-102 dataset · Inception score · Human rank This study is funded by the General Program of the National Natural Science Foundation of China (No: 61977029).

本校为第一且通讯机构

Synthesizing high-resolution realistic images from text description using one iteration Generative Adversarial Network (GAN) is difficult without using any additional techniques because mostly the blurry artifacts and mode collapse problems are occurring. To reduce these problems, this paper proposes an Iterative Generative Adversarial Network (iGAN) which takes three iterations to synthesize high-resolution realistic image from their text description. In the \(1^{st}\) iteration, GAN synthesizes a low-resolution \(64 \times 64\) pixels basic shape and basic color image from the text descripti...