Multiple Heads are Better than One: Few-shot Font Generation with Multiple Localized Experts

Park, Song; Chun, Sanghyuk; Cha, Junbum; Lee, Bado; Shim, Hyunjung

doi:10.1109/iccv48922.2021.01364

Cited by 35 publications

(18 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We used the same training and testing datasets in Section 4.1 for all methods, and for the other settings, we followed the original settings of the papers [ 12 , 13 ]. We omit component-wise encoding approaches [ 15 , 20 , 21 ] from the comparison group because they need additional labels.…”

Section: Methodsmentioning

confidence: 99%

“…Some studies [ 15 , 20 , 21 ] did not encode the entire image to extract style information but, instead, extracted style information for each component that makes up a syllable and used it to generate fonts. However, this requires additional labels for components and is not suitable for such languages with a large number of components such as Chinese.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Arbitrary Font Generation by Encoder Learning of Disentangled Features

Lee

Baek

Choi

2022

Sensors

View full text Add to dashboard Cite

Making a new font requires graphical designs for all base characters, and this designing process consumes lots of time and human resources. Especially for languages including a large number of combinations of consonants and vowels, it is a heavy burden to design all such combinations independently. Automatic font generation methods have been proposed to reduce this labor-intensive design problem. Most of the methods are GAN-based approaches, and they are limited to generate the trained fonts. In some previous methods, they used two encoders, one for content, the other for style, but their disentanglement of content and style is not sufficiently effective in generating arbitrary fonts. Arbitrary font generation is a challenging task because learning text and font design separately from given font images is very difficult, where the font images have both text content and font style in each image. In this paper, we propose a new automatic font generation method to solve this disentanglement problem. First, we use two stacked inputs, i.e., images with the same text but different font style as content input and images with the same font style but different text as style input. Second, we propose new consistency losses that force any combination of encoded features of the stacked inputs to have the same values. In our experiments, we proved that our method can extract consistent features of text contents and font styles by separating content and style encoders and this works well for generating unseen font design from a small number of reference font images that are human-designed. Comparing to the previous methods, the font designs generated with our method showed better quality both qualitatively and quantitatively than those with the previous methods for Korean, Chinese, and English characters. e.g., 17.84 lower FID in unseen font compared to other methods.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Arbitrary Font Generation by Encoder Learning of Disentangled Features

Lee

Baek

Choi

2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Raster font generation has been widely studied [1,2,3,4]. However, attempts to generate vector fonts have been limited, which we describe below.…”

Section: Related Workmentioning

confidence: 99%

“…Designing Chinese vector fonts is generally a time-consuming and labor-intensive task, and thus there is a demand for automatic font generation to support creators. While the generation of Chinese fonts as raster images [1,2,3,4] has been widely studied, the generation of fonts as vector graphics, which is the original form of fonts, has been barely tackled yet. Scale-invariance of vector fonts is an essential property of fonts for use at various scales, from documents to posters, and for this reason, most of the fonts commonly used in the real world are in vector format.…”

Section: Introductionmentioning

confidence: 99%

SVG Vector Font Generation for Chinese Characters with Transformer

Aoki¹,

Aizawa²

2022

Preprint

View full text Add to dashboard Cite

Designing fonts for Chinese characters is highly labor-intensive and time-consuming. While the latest methods successfully generate the English alphabet vector font, despite the high demand for automatic font generation, Chinese vector font generation has been an unsolved problem owing to its complex shape and numerous characters. This study addressed the problem of automatically generating Chinese vector fonts from only a single style and content reference. We proposed a novel network architecture with Transformer and loss functions to capture structural features without differentiable rendering. Although the dataset range was still limited to the sans-serif family, we successfully generated the Chinese vector font for the first time using the proposed method.

show abstract

“…Shimoda et al [8] suggested a text vectorization technique by leveraging the advantage of differentiable text rendering to accurately reproduce the input raster text in a resolutionfree parametric format. On the other hand, the printed character image generation [42]- [48] has been comprehensively studied. These methods simulate writing styles such as fonts, thickness, and artistic style while preserving the base structure.…”

Section: Text Image Generationmentioning

confidence: 99%

APRNet: Attention-based Pixel-wise Rendering Network for Photo-Realistic Text Image Generation

Shi¹,

Ding²,

Chen³

et al. 2022

Preprint

View full text Add to dashboard Cite

Style-guided text image generation tries to synthesize text image by imitating reference image's appearance while keeping text content unaltered. The text image appearance includes many aspects. In this paper, we focus on transferring style image's background and foreground color patterns to the content image to generate photo-realistic text image. To achieve this goal, we propose 1) a content-style cross attention based pixel sampling approach to roughly mimicking the style text image's background; 2) a pixel-wise style modulation technique to transfer varying color patterns of the style image to the content image spatial-adaptively; 3) a cross attention based multi-scale style fusion approach to solving text foreground misalignment issue between style and content images; 4) an image patch shuffling strategy to create style, content and ground truth image tuples for training. Experimental results on Chinese handwriting text image synthesis with SCUT-HCCDoc and CASIA-OLHWDB datasets demonstrate that the proposed method can improve the quality of synthetic text images and make them more photo-realistic.

show abstract

Multiple Heads are Better than One: Few-shot Font Generation with Multiple Localized Experts

Cited by 35 publications

References 30 publications

Arbitrary Font Generation by Encoder Learning of Disentangled Features

Arbitrary Font Generation by Encoder Learning of Disentangled Features

SVG Vector Font Generation for Chinese Characters with Transformer

APRNet: Attention-based Pixel-wise Rendering Network for Photo-Realistic Text Image Generation

Contact Info

Product

Resources

About