Image generation step by step: animation generation-image translation

被引:5
作者
Jing, Beibei [1 ]
Ding, Hongwei [1 ]
Yang, Zhijun [1 ]
Li, Bo [1 ]
Liu, Qianlin [1 ]
机构
[1] Yunnan Univ, Sch Informat Sci & Engn, Kunming, Yunnan, Peoples R China
基金
中国国家自然科学基金;
关键词
Generative adversarial networks; LGAN (Link Generative Adversarial Networks); Unconditional generation part; Anime images conditional generation part; Super-resolution network; WASSERSTEIN DISTANCE;
D O I
10.1007/s10489-021-02835-z
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Generative adversarial networks play an important role in image generation, but the successful generation of high-resolution images from complex data sets remains a challenging goal. In this paper, we propose the LGAN (Link Generative Adversarial Networks) model, which can effectively enhance the quality of the synthesized images. The LGAN model consists of two parts, G1 and G2. G1 is responsible for the unconditional generation part, which generates anime images with highly abstract features containing few coefficients but continuous image elements covering the overall image features. Moreover, G2 is responsible for the conditional generation part (image translation), consisting of mapping and Superresolution networks. The mapping network fills the output of G1 into the real-world image after semantic segmentation or edge detection processing; the Superresolution network super-resolves the actual picture after completing mapping to improve the image's resolution. In the comparison test with WGAN, SAGAN, WGAN-GP and PG-GAN, this paper's LGAN(SEG) leads 64.36 and 12.28, respectively, fully proving the model's superiority.
引用
收藏
页码:8087 / 8100
页数:14
相关论文
共 50 条
  • [31] Panoramic Image Generation: From 2-D Sketch to Spherical Image
    Duan, Yiping
    Han, Chaoyi
    Tao, Xiaoming
    Geng, Bingrui
    Du, Yunfei
    Lu, Jianhua
    [J]. IEEE JOURNAL OF SELECTED TOPICS IN SIGNAL PROCESSING, 2020, 14 (01) : 194 - 208
  • [32] Generative image completion with image-to-image translation
    Shuzhen Xu
    Qing Zhu
    Jin Wang
    [J]. Neural Computing and Applications, 2020, 32 : 7333 - 7345
  • [33] Generative image completion with image-to-image translation
    Xu, Shuzhen
    Zhu, Qing
    Wang, Jin
    [J]. NEURAL COMPUTING & APPLICATIONS, 2020, 32 (11) : 7333 - 7345
  • [34] Hypercomplex Image-to-Image Translation
    Grassucci, Eleonora
    Sigillo, Luigi
    Uncini, Aurelio
    Comminiello, Danilo
    [J]. 2022 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2022,
  • [35] Multi-branch Semantic GAN for Infrared Image Generation from Optical Image
    Li, Lei
    Li, Pengfei
    Yang, Meng
    Gao, Shibo
    [J]. INTELLIGENCE SCIENCE AND BIG DATA ENGINEERING: VISUAL DATA ENGINEERING, PT I, 2019, 11935 : 484 - 494
  • [36] Sequential Image Synthesis for Human Activity Video Generation
    Khan, Fahim Hasan
    de Silva, Akila
    Yetukuri, Jayanth
    Norouzi, Narges
    [J]. IMAGE ANALYSIS AND RECOGNITION (ICIAR 2019), PT II, 2019, 11663 : 129 - 133
  • [37] GAN-BASED SYNTHETIC GASTROINTESTINAL IMAGE GENERATION
    Adjei, Prince E.
    Lonsek, Zenebe M.
    Rao, Nini
    [J]. 2020 17TH INTERNATIONAL COMPUTER CONFERENCE ON WAVELET ACTIVE MEDIA TECHNOLOGY AND INFORMATION PROCESSING (ICCWAMTIP), 2020, : 338 - 342
  • [38] Evaluation of Modern Generative Networks for EchoCG Image Generation
    Rakhmetulayeva, Sabina
    Zhanabekov, Zhandos
    Bolshibayeva, Aigerim
    [J]. CMC-COMPUTERS MATERIALS & CONTINUA, 2024, 81 (03): : 4503 - 4523
  • [39] Image generation with self pixel-wise normalization
    Yeo, Yoon-Jae
    Sagong, Min-Cheol
    Park, Seung
    Ko, Sung-Jea
    Shin, Yong-Goo
    [J]. APPLIED INTELLIGENCE, 2023, 53 (08) : 9409 - 9423
  • [40] Segmentation mask-guided person image generation
    Meichen Liu
    Xin Yan
    Chenhui Wang
    Kejun Wang
    [J]. Applied Intelligence, 2021, 51 : 1161 - 1176