Abstract: Content and style disentanglement is an effective way to achieve few-shot font generation. It allows to transfer the style of the font image in a source domain to the style defined with a few reference images in a target domain. However, the content feature extracted using a representative font might not be optimal. In light of this, we propose a content fusion module (CFM) to project the content feature into a linear space defined by the content features of basis fonts, which can take the variation of content features caused by different fonts into consideration. Our method also allows to optimize the style representation vector of reference images through a lightweight iterative style-vector refinement (ISR) strategy. Moreover, we treat the 1D projection of a character image as a probability distribution and leverage the distance between two distributions as the reconstruction loss (namely projected character loss, PCL). Compared to L2 or L1 reconstruction loss, the distribution distance pays more attention to the global shape of characters. We have evaluated our method on a dataset of 300 fonts with 6.5k characters each. Experimental results verify that our method outperforms existing state-of-the-art few-shot font generation methods by a large margin. The source code can be found at https://github.com/wangchi95/CF-Font.

Abstract: Automatic generation of fonts can be an important aid to typeface design. Many current approaches regard glyphs as pixelated images, which present artifacts when scaling and inevitable quality losses after vectorization. On the other hand, existing vector font synthesis methods either fail to represent the shape concisely or require vector supervision during training. To push the quality of vector font synthesis to the next level, we propose a novel dual-part representation for vector glyphs, where each glyph is modeled as a collection of closed “positive” and “negative” path pairs. The glyph contour is then obtained by boolean operations on these paths. We first learn such a representation only from glyph images and devise a subsequent contour refinement step to align the contour with an image representation to further enhance details. Our method, named DualVector, outperforms state-of-the-art methods in vector font synthesis both quantitatively and qualitatively. Our synthesized vector fonts can be easily converted to common digital font formats like TrueType Font for practical use. The code is released at https://github.com/thuliu-yt16/dualvector.

Abstract: Automatic few-shot font generation (AFFG), aiming at generating new fonts with only a few glyph references, reduces the labor cost of manually designing fonts. However, the traditional AFFG paradigm of style-content disentanglement cannot capture the diverse local details of different fonts. So, many component-based approaches are proposed to tackle this problem. The issue with component-based approaches is that they usually require special pre-defined glyph components, e.g., strokes and radicals, which is infeasible for AFFG of different languages. In this paper, we present a novel font generation approach by aggregating styles from character similarity-guided global features and stylized component-level representations. We calculate the similarity scores of the target character and the referenced samples by measuring the distance along the corresponding channels from the content features, and assigning them as the weights for aggregating the global style features. To better capture the local styles, a cross-attention-based style transfer module is adopted to transfer the styles of reference glyphs to the components, where the components are self-learned discrete latent codes through vector quantization without manual definition. With these designs, our AFFG method could obtain a complete set of component-level style representations, and also control the global glyph characteristics. The experimental results reflect the effectiveness and generalization of the proposed method on different linguistic scripts, and also show its superiority when compared with other state-of-the-art methods. The source code can be found at https://github.com/awei669/VQ-Font.

Abstract: ABSTRACT Tag weight differences in tag maps are usually reflected by different font sizes. With this strategy, low weighted tags may be ignored and tag sizes may be misjudged due to differing word length, character height and word width. To address these shortcomings, this paper improves the layout method of tag maps by presetting anchor points. We construct weight contours based on corner points and center points of tag outer envelope rectangles and adopt hypsometric tinting as background color to reflect tag weight differences. Finally, we compare and analyze the background color strategy with the font size strategy from the perspectives of tag selection, tag recognition, tag recall, confidence, readability, and preference through eye movement experiments and questionnaire surveys. The results show that both strategies exhibit advantages (when tags are devoid of semantics), with the font size strategy being favored slightly in this abstract case. We provide tag map designers with a new visualization scheme for the expression of tag weight.

Abstract: Variable font file technology enables adjusting fonts on scaled axes that can include weight, and grade. While making text bold increases the character width, grade achieves boldness without increasing character width or causing text reflow. Through two studies with a total of 459 participants, we examined the effect of varying grade levels on both glancing and paragraph reading tasks in light and dark modes. We show that dark text on a light background (Light Mode) is read reliably faster than its polar opposite (Dark Mode). We found an effect of mode for both glance and paragraph reading and an effect of grade for LM with heavier, increased grade levels. Paragraph readers are not choosing, or preferring, LM over DM despite fluency benefits and reported visual clarity. Software designers can vary grade across the tested font formats to influence design aesthetics and user preferences without worrying about reducing reading fluency.

Abstract: Font design is of vital importance in the digital content design and modern printing industry. Developing algorithms capable of automatically synthesizing vector fonts can significantly facilitate the font design process. However, existing methods mainly concentrate on raster image generation, and only a few approaches can directly synthesize vector fonts. This paper proposes an end-to-end trainable method, VecFontSDF, to reconstruct and synthesize high-quality vector fonts using signed distance functions (SDFs). Specifically, based on the proposed SDF-based implicit shape representation, VecFontSDF learns to model each glyph as shape primitives enclosed by several parabolic curves, which can be precisely converted to quadratic Bézier curves that are widely used in vector font products. In this manner, most image generation methods can be easily extended to synthesize vector fonts. Qualitative and quantitative experiments conducted on a publicly-available dataset demonstrate that our method obtains high-quality results on several tasks, including vector font reconstruction, interpolation, and few-shot vector font synthesis, markedly outperforming the state of the art.

Abstract: Hiding information in text documents has been a hot topic recently, with the most typical schemes of utilizing fonts. By constructing several fonts with similar appearances, information can be effectively represented and embedded in documents. However, due to the unstructured characteristic, font vectors are more difficult to synthesize than font images. Existing methods mainly use handcrafted features to design the fonts manually, which is time-consuming and labor-intensive. Moreover, due to the diversity of fonts, handcrafted features are not generalizable to different fonts. Besides, in practice, since documents might be distorted through transmission, ensuring extractability under distortions is also an important requirement. Therefore, three requirements are imposed on vector font generation in this domain: automaticity, generalizability, and robustness. However, none of the existing methods can satisfy these requirements well and simultaneously. To satisfy the above requirements, we propose AutoStegaFont, an automatic vector font synthesis scheme for hiding information in documents. Specifically, we design a two-stage and dual-modality learning framework. In the first stage, we jointly train an encoder and a decoder to invisibly encode the font images with different information. To ensure robustness, we target designing a noise layer to work with the encoder and decoder during training. In the second stage, we employ a differentiable rasterizer to establish a connection between the image and the vector modality. Then, we design an optimization algorithm to convey the information from the encoded image to the corresponding vector. Thus the encoded font vectors can be automatically generated. Extensive experiments demonstrate the superior performance of our scheme in automatically synthesizing vector fonts for hiding information in documents, with robustness to distortions caused by low-resolution screenshots, printing, and photography. Besides, the proposed framework has better generalizability to fonts with diverse styles and languages.

Abstract: There are various font styles in the world. Different styles give different impressions and readability. This paper analyzes the relationship between font styles and contextual factors that might affect font style selection with large-scale datasets. For example, we will analyze the relationship between font style and its surrounding object (such as “bus”) by using about 800,000 words in the Open Images dataset. We also use a book cover dataset to analyze the relationship between font styles with book genres. Moreover, the meaning of the word is assumed as another contextual factor. For these numeric analyses, we utilize our own font-style feature extraction model and word2vec. As a result of co-occurrence-based relationship analysis, we found several instances of specific font styles being used for specific contextual factors.

Abstract: The essence of font style transfer is to move the style features of an image into a font while maintaining the font’s glyph structure. At present, generative adversarial networks based on convolutional neural networks play an important role in font style generation. However, traditional convolutional neural networks that recognize font images suffer from poor adaptability to unknown image changes, weak generalization abilities, and poor texture feature extractions. When the glyph structure is very complex, stylized font images cannot be effectively recognized. In this paper, a deep deformable style transfer network is proposed for artistic font style transfer, which can adjust the degree of font deformation according to the style and realize the multiscale artistic style transfer of text. The new model consists of a sketch module for learning glyph mapping, a glyph module for learning style features, and a transfer module for a fusion of style textures. In the glyph module, the Deform-Resblock encoder is designed to extract glyph features, in which a deformable convolution is introduced and the size of the residual module is changed to achieve a fusion of feature information at different scales, preserve the font structure better, and enhance the controllability of text deformation. Therefore, our network has greater control over text, processes image feature information better, and can produce more exquisite artistic fonts.

Abstract: Abstract In this paper, we propose a cross-language font style transfer system that can synthesize a new font by observing only a few samples from another language. Automatic font synthesis is a challenging task and has attracted much research interest. Most previous works addressed this problem by transferring the style of the given subset to the content of unseen ones. Nevertheless, they only focused on the font style transfer in the same language. In many cases, we need to learn font style from one language and then apply it to other languages. Existing methods make this difficult to accomplish because of the abstraction of style and language differences. To address this problem, we specifically designed the network into a multi-level attention form to capture both local and global features of the font style. To validate the generative ability of our model, we constructed an experimental font dataset of 847 fonts, each containing English and Chinese characters with the same style. Results show that our model generates 80.3% of users’ preferred images compared with state-of-the-art models.

Abstract: Mathematics is presented in a variety of font types across materials (e.g., textbooks, online problems); however, little is known about the effects of font type on students’ mathematical performance. Undergraduate students (N = 121) completed three mathematical tasks in a one-hour online session in one of three font conditions: Times New Roman (n = 45), Kalam (n = 41), or handwriting (n = 35). We examined whether font type impacted students’ performance, as measured by accuracy and response time, on the Perceptual Math Equivalence Task, error identification task, and equation-solving task. Compared to students in the Kalam and handwriting conditions, students in the Times New Roman condition were less accurate on the Perceptual Math Equivalence Task in which they judged whether two expressions were equivalent or not equivalent. We did not find differences between conditions in performance on error identification and equation-solving tasks. The findings have implications for research and practice. Specifically, researchers and educators may choose font types in which they present mathematics information with consideration, as font types may impact students’ mathematical processing and performance.

Abstract: Abstract. The media of Islamic organizations in Indonesia can now be said to be an online media that is needed by the community, especially those who are involved in these Islamic organizations. The presence of information that feels closer personally and geographically makes Islamic mass media media the community's choice of media when the need for valid information is difficult to find. Of the various types of Islamic media organizations in Indonesia, each has its own characteristics, ranging from the variety of websites they manage to different writing styles, one of which is the style of the environment. This research focuses on how the diversity of writing between NU Jabar and Persis media can be written so as to produce the finding that how a media can determine the language style / style of its environment which is so diverse, even though it is still in the same source or rule, namely Islam. The method used in this research is qualitative with a case study approach, and uses data collection techniques in the form of observations, interviews, and literature studies. The results of the research are 1) The writing format on the West Java and Persis NU media there are some basic differences, such as the difference in the font used where NU uses the Metropolis Semibold Family font, while Persis.or.id uses the Trebuchet MS font. In the West Java NU media, they set a standard number of words for each article, which is 300 words, while in Persis they are more flexible by not specifying a minimum number of words. 2) There are differences in the mention of religious figures in the NU Jabar and Persis media, NU Jabar uses the words H. (Haji) and KH. (Kyai Haji) for the designation of religious figures, while Persis mentions Ustad or Al-Ustad. There are also many differences regarding the mention of events, for example, the mention of the word Ramadan, Prayer, Pupils and so on. 3) Basically, the West Java and Persis NU media still refer to the KBBI in determining each word choice they use. However, both of them agreed that some differences in words or what is commonly called the selingkung style is a natural thing considering this is the hallmark of an organization which eventually becomes the hallmark of the media. Abstrak. Media ormas Islam di Indonesia saat ini bisa dibilang menjadi media online yang dibutuhkan oleh masyarakat, terutama mereka yang berkecimpung di organisasi Islam tersebut. Hadirnya informasi yang terasa lebih dekat secara personal maupun geografis menjadikan media ormas Islam menjadi media pilihan masyarakat ketika kebutuhan akan informasi yang valid sulit ditemukan. Dari berbagai jenis organisasi media Islam di Indonesia, masing-masing memiliki ciri khasnya tersendiri, mulai dari keberagaman website yang mereka kelola hingga gaya penulisan yang berbeda-beda, salah satunya adalah gaya selingkungnya. Penelitian ini berfokus kepada bagaimana keberagaman penulisan antara media NU Jabar dan Persis dapat ditulis sehingga menghasilkan temuan bahwa bagaimana sebuah media bisa menentukan gaya bahasa/gaya selingkungnya yang begitu beragam, walaupun masih dalam sumber atau kaidah yang sama yaitu agama Islam. Metode yang digunakan pada penelitian ini yaitu kualitatif dengan pendekatan studi kasus, serta menggunakan teknik pengumpulan data berupa observasi, wawancara, dan studi pustaka. Adapun hasil penelitiannya yaitu 1) Format penulisan pada media NU Jabar dan Persis terdapat beberapa perbedaan yang mendasar, seperti perbedaan font yang digunakan dimana NU menggunakan font MetropolisSemibold Family, sedangkan Persis.or.id menggunakan font Trebuchet MS. Pada media NU Jabar, mereka menetapkan standar jumlah kata tiap tulisan yaitu 300 kata, sedangkan pada media Persis mereka lebih fleksibel dengan tidak menentukan minimal jumlah kata. 2) Terdapat perbedaan penyebutan tokoh keagamaan pada media NU Jabar dan Persis, NU Jabar menggunakan kata H. (Haji) dan KH. (Kyai Haji) untuk sebutan tokoh keagamaan, sedangkan Persis menyebutkan Ustad atau Al-Ustad. Terdapat pula banyak perbedaan terkait penyebutan peristiwa contohnya seperti penyebutan kata Ramadhan, Sholat, Murid dan sebagainya. 3) Pada dasarnya, media NU Jabar dan Persis masih merujuk kepada KBBI dalam menentukan setiap pemilihan kata yang mereka gunakan. Namun keduanya bersepakat bahwa beberapa perbedaan kata atau yang biasa disebut gaya selingkung merupakan hal yang wajar mengingat hal ini merupakan ciri khas suatu organisasi yang akhirnya menjadi ciri khas pada media tersebut.

Abstract: Chinese calligraphy, revered globally for its therapeutic and mindfulness benefits, encompasses styles such as Regular (Kai Shu), Running (Xing Shu), Official (Li Shu), and Cursive (Cao Shu) scripts. Beginners often start with Regular script, advancing to more intricate styles like Cursive. Each style, marked by unique historical calligraphers' contributions, requires learners to discern distinct nuances. The integration of AI in calligraphy analysis, collection, recognition, and classification are pivotal. This study introduces an innovative Convolutional Neural Network (CNN) architecture, pioneering the application of CNN in the classification of Chinese calligraphy. Focusing on the four principal calligraphers' styles from the Tang dynasty (690-907 A.D), this research spotlights the era when the traditional regular script font (Kai Shu) was refined. A comprehensive dataset of 8282 samples from these calligraphers, representing the zenith of regular style, was compiled for CNN training and testing. The model distinguishes personal styles for classification, showing superior performance over existing networks. Achieving 89.5-96.2% accuracy in calligraphy classification, our approach underscores the significance of CNN in both font and artistic style categorization. This research paves the way for advanced studies in Chinese calligraphy and its cultural implications.

Abstract: Reading comprehension is an essential skill. It is unclear whether and to what degree typography and font personalization may impact reading comprehension in younger readers. With advancements in technology, it is now feasible to personalize digital reading formats in general technology tools, but this feature is not yet available for many educational tools. The current study aimed to investigate the effect of character width and inter-letter spacing on reading speed and comprehension. We enrolled 94 children (kindergarten–8th grade) and compared performance with six font variations on a word-level semantic decision task (Experiment 1) and a passage-level comprehension task (Experiment 2). Reading speed and comprehension were analyzed using generalized linear mixed-effects regression models. Independent samples t-tests compared speed and comprehension accuracy on personal best vs. worst font variation. A stability analysis was conducted to determine if participants had a stable personal best font variation within Experiment 1. The Experiment 1 stability analysis was statistically significant, and 58% of participants had a stable personal best font variation. Personal best font variations yielded significantly higher comprehension accuracy in both Experiments 1 and 2 and faster reading in Experiment 2. Using digital technology to personalize font may have important implications for school-aged readers.

Abstract: The optical character recognition for the right to left and cursive languages such as Arabic is challenging and received little attention from researchers in the past compared to the other Latin languages. Moreover, the absence of a standard publicly available dataset for several low-resource languages, including the Pashto language remained a hurdle in the advancement of language processing. Realizing that, a clean dataset is the fundamental and core requirement of character recognition, this research begins with dataset generation and aims at a system capable of complete language understanding. Keeping in view the complete and full autonomous recognition of the cursive Pashto script. The first achievement of this research is a clean and standard dataset for the isolated characters of the Pashto script. In this paper, a database of isolated Pashto characters for forty four alphabets using various font styles has been introduced. In order to overcome the font style shortage, the graphical software Inkscape has been used to generate sufficient image data samples for each character. The dataset has been pre-processed and reduced in dimensions to 32 × 32 pixels, and further converted into the binary format with a black background and white text so that it resembles the Modified National Institute of Standards and Technology (MNIST) database. The benchmark database is publicly available for further research on the standard GitHub and Kaggle database servers both in pixel and Comma Separated Values (CSV) formats.

Abstract: ABLY Creative Font adalah marketplace yang menyediakan jasa Creative Font yang saat ini belum memiliki standarisasi penilaian dalam pemilihan calon Font Authors terbaik yang layak diterima. Proses penerimaan calon Font Authors hanya didasarkan pada penilaian individual dari founder ABLY Creative Font yang menyeleksi berkas secara langsung. Hal ini berimbas pada kualitas Font Authors yang diterima belum sesuai dengan keinginan ABLY Creative Font. Untuk itu dibutuhkan sebuah sistem yang dapat menyajikan penilaian yang objektif dalam penentuan calon Font Authors terbaik. Sistem Pendukung Keputusan (SPK) merupakan sistem berbasis komputer yang dipakai untuk mendukung pengambilan keputusan dalam suatu organisasi, instansi ataupun perusahaan. Metode MOORA (Multi-Objective Optimization On The Basis Of Ratio Analysis) merupakan metode multi objektif sistem yang mengoptimalkan dua atau lebih atribut yang saling bertentangan secara bersamaan. Hasil penerapan SPK metode MOORA memberikan penilaian yang objektif berdasarkan kriteria penilaian penentuan kelayakan penerimaan Font Authors terbaik yang akan diterima di ABLY Creatif Font.

Abstract: Abstract This study examines whether the common typographical recommendations for texts in easy-to-read language are helpful. So far, using a sans-serif font in 14 pt has been suggested. The statistical evaluation shows that Thesis TheSans is read significantly faster than Arial. Fonts with serifs are read slightly faster. The font size of 12 pt is large enough. The study clearly indicates that the strict commitment to sans serif fonts—and the Arial font in particular—for texts in easy-to-read language is wrong. The results can be used by designers and clients as a valid decision-making aid when creating barrier-free communication.

Abstract: Currently, many Chinese font translation models adopt the method of dividing character components to improve the quality of generated font images. However, character components require a large amount of manual annotation to decompose characters and determine the composition of each character as input for training. In this paper, we establish a Chinese font translation model based on generative adversarial network without decomposition. First, we improve the method of image enhancement for Chinese character images. It helps the model learning structure information of Chinese character strokes to generate font images with complete and accurate strokes. Second, we propose a visual attention adversarial network. By using visual attention block, the network catches global and local features for constructing details of characters. Experiments demonstrate our method generates high-quality Chinese character images with great style diversity including calligraphy characters.

Abstract: With the increase of various media, fonts continue to be newly developed. In Korea, numerous ‘Hangul’ fonts are also being developed, and as a result, the need for research on determining the similarity between fonts is emerging. For example, when creating a document, the font to be used must be downloaded from each computing environment. However, this is a very cumbersome process. If there is a font that is not supported in the system, the above problem can be easily solved by recommending the most similar font that can replace it. According to this need, we conducted various prior studies for similar font recommendations. As a result, we developed a ‘stroke element’ that exists in each consonant and vowel in Korean font and developed a font recommendation model using a stroke element. However, there is a limitation in that the existing research was studied only for the structured fonts corresponding to the printed type. Additionally, the font size was not considered in the font recommendation. In this study, two experiments were conducted to expand the font recommendation model by supplementing the limitations of existing studies. First, in order to enable similar font recommendations based on the stroke element even in fonts with various shapes, the font was classified according to the shape, and the stroke elements in each classification were detected. Second, when the font sizes were different, the change in the font recommendations result based on the stroke element was analyzed. In conclusion, we found that it was necessary to find a plan to extract stroke elements for font recommendation of fonts that do not belong to standard fonts. In addition, since the influence of the stroke element varies depending on the size of the font, we propose a stroke element weight model that can be used for recommendation by reflecting it.