Few-shot Font Generation with Localized Style Representations and Factorization

被引:0
|
作者
Park, Song [1 ,3 ]
Chun, Sanghyuk [2 ,3 ]
Cha, Junbum [3 ]
Lee, Bado [3 ]
Shim, Hyunjung [1 ]
机构
[1] Yonsei Univ, Sch Integrated Technol, Seoul, South Korea
[2] NAVER AI LAB, Seoul, South Korea
[3] NAVER CLOVA, Seoul, South Korea
基金
新加坡国家研究基金会;
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Automatic few-shot font generation is a practical and widely studied problem because manual designs are expensive and sensitive to the expertise of designers. Existing few-shot font generation methods aim to learn to disentangle the style and content element from a few reference glyphs, and mainly focus on a universal style representation for each font style. However, such approach limits the model in representing diverse local styles, and thus makes it unsuitable to the most complicated letter system, e.g., Chinese, whose characters consist of a varying number of components (often called "radical") with a highly complex structure. In this paper, we propose a novel font generation method by learning localized styles, namely component-wise style representations, instead of universal styles. The proposed style representations enable us to synthesize complex local details in text designs. However, learning component-wise styles solely from reference glyphs is infeasible in the few-shot font generation scenario, when a target script has a large number of components, e.g., over 200 for Chinese. To reduce the number of reference glyphs, we simplify component-wise styles by a product of component factor and style factor, inspired by low-rank matrix factorization. Thanks to the combination of strong representation and a compact factorization strategy, our method shows remarkably better few-shot font generation results (with only 8 reference glyph images) than other state-of-the-arts, without utilizing strong locality supervision, e.g., location of each component, skeleton, or strokes. The source code is available at https://github.com/clovaai/lffont.
引用
收藏
页码:2393 / 2402
页数:10
相关论文
共 50 条
  • [41] DataDream: Few-Shot Guided Dataset Generation
    Kim, Jae Myung
    Bader, Jessica
    Alaniz, Stephan
    Schmid, Cordelia
    Akata, Zeynep
    COMPUTER VISION - ECCV 2024, PT LXXI, 2025, 15129 : 252 - 268
  • [42] FEW-SHOT GENERATION BY MODELING STEREOSCOPIC PRIORS
    Wang, Yuehui
    Wang, Qing
    Zhang, Dongyu
    2022 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2022, : 2370 - 2374
  • [43] Counterfactual Generation Framework for Few-Shot Learning
    Dang, Zhuohang
    Luo, Minnan
    Jia, Chengyou
    Yan, Caixia
    Chang, Xiaojun
    Zheng, Qinghua
    IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2023, 33 (08) : 3747 - 3758
  • [44] Few-shot Fish Image Generation and Classification
    Guo, Zonghui
    Zhang, Liqiang
    Jiang, Yufeng
    Niu, Wenjie
    Gu, Zhaorui
    Zheng, Haiyong
    Wang, Guoyu
    Zheng, Bing
    GLOBAL OCEANS 2020: SINGAPORE - U.S. GULF COAST, 2020,
  • [45] Few-Shot Few-Shot Learning and the role of Spatial Attention
    Lifchitz, Yann
    Avrithis, Yannis
    Picard, Sylvaine
    2020 25TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), 2021, : 2693 - 2700
  • [46] HFH-Font: Few-shot Chinese Font Synthesis with Higher Quality, Faster Speed, and Higher Resolution
    Li, Hua
    Lian, Zhouhui
    ACM TRANSACTIONS ON GRAPHICS, 2024, 43 (06):
  • [47] Layer similarity guiding few-shot Chinese style transfer
    Li, Yumei
    Lin, Guangfeng
    He, Menglan
    Yuan, Dan
    Liao, Kaiyang
    VISUAL COMPUTER, 2024, 40 (04): : 2265 - 2278
  • [48] Layer similarity guiding few-shot Chinese style transfer
    Yumei Li
    Guangfeng Lin
    Menglan He
    Dan Yuan
    Kaiyang Liao
    The Visual Computer, 2024, 40 : 2265 - 2278
  • [49] Master: Meta Style Transformer for Controllable Zero-Shot and Few-Shot Artistic Style Transfer
    Tang, Hao
    Liu, Songhua
    Lin, Tianwei
    Huang, Shaoli
    Li, Fu
    He, Dongliang
    Wang, Xinchao
    2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2023, : 18329 - 18338
  • [50] Learn to aggregate global and local representations for few-shot learning
    Abdelaziz, Mounir
    Zhang, Zuping
    MULTIMEDIA TOOLS AND APPLICATIONS, 2023, 82 (21) : 32991 - 33014