Meta tags:
keywords=DS-Fusion, Typogrphy, Artistic, Diffusion Models, Semantic Typography, AI Art;
description=DS-Fusion: create artistic typography automatically;

Headings (most frequently used words):

stylization, ds, fusion, artistic, typography, letter, via, discriminated, and, stylized, diffusion, create, automatically, abstract, video, pipeline, single, multi, font, specific, attribute,

Text of the page (most frequently used words):
the (37), and (13), style (9), #fusion (8), input (7), stylization (7), #letter (7), artistic (7), diffusion (7), word (5), typography (5), stylized (5), for (4), font (4), more (4), with (4), glyph (4), model (4), code (3), results (3), letters (3), single (3), generated (3), images (3), latent (3), discriminator (3), our (3), discriminated (3), website (2), source (2), based (2), creative (2), use (2), attribute (2), produced (2), fonts (2), glyphs (2), shape (2), including (2), one (2), convey (2), color (2), which (2), pipeline (2), are (2), utilizes (2), process (2), space (2), given (2), module (2), models (2), denoising (2), generator (2), samples (2), output (2), automatically (2), via (2), nerfies, this, licensed, under, commons, attribution, sharealike, international, license, attributes, prompts, can, effective, means, further, fine, tune, figures, below, shows, varying, dragon, maintaing, stroke, thickness, slanting, accents, specific, synthesizing, typographies, from, multiple, challenging, due, added, structural, complexities, inputs, well, global, context, account, ways, than, demonstrates, its, ability, utilize, all, semantic, features, manner, multi, show, compelling, each, result, composing, renderings, remaining, render, other, same, select, complementing, often, dominant, takes, prompt, image, according, first, construct, then, introduces, blend, into, parameters, pre, trained, frozen, there, icon, lock, bottom, right, denotes, iterative, noise, injection, video, coined, address, assortment, challenges, task, hand, conflicting, goals, legibility, lack, ground, truth, immense, search, approach, large, language, bridge, texts, visual, build, generative, backbone, specifically, employ, ldm, key, addition, cnn, adapt, onto, text, uses, rasterized, real, fake, unsupervised, introduce, novel, method, generate, visually, semantics, while, ensuring, that, remains, readable, stylizing, abstract, create, demo, beta, supplementary, paper, simon, fraser, university, hao, zhang, ali, mahdavi, amiri, yizhi, wang, maham, tanveer,

Text of the page (random words):
ds fusion artistic typography via discriminated and stylized diffusion ds fusion artistic typography via discriminated and stylized diffusion maham tanveer yizhi wang ali mahdavi amiri hao zhang simon fraser university paper supplementary demo beta code ds fusion create artistic typography automatically abstract we introduce a novel method to automatically generate an artistic typography by stylizing one or more letter fonts to visually convey the semantics of an input word while ensuring that the output remains readable to address an assortment of challenges with our task at hand including conflicting goals artistic stylization vs legibility lack of ground truth and immense search space our approach utilizes large language models to bridge texts and visual images for stylization and build an unsupervised generative model with a diffusion model backbone specifically we employ the denoising generator in latent diffusion model ldm with the key addition of a cnn based discriminator to adapt the input style onto the input text the discriminator uses rasterized images of a given letter word font as real samples and output of the denoising generator as fake samples our model is coined ds fusion for discriminated and stylized diffusion video pipeline the pipeline of ds fusion which takes as input a style prompt and a glyph image the style images are generated according to the style word and attribute ds fusion first utilizes a latent diffusion process to construct the latent space of the given style and then introduces a discriminator to blend the style into the glyph shape the parameters of a module are pre trained and frozen if there is an icon of a lock on the bottom right the module denotes the iterative noise injection process of diffusion models single letter stylization we show compelling single letter artistic typography results generated by ds fusion each result is produced by composing generated single letter results with renderings of the remaining letters in the input style word to render the other letters we use the same font as the stylized glyph and select a complementing color which is often the dominant color in the stylized glyph multi letter stylization synthesizing artistic typographies from multiple glyphs is more challenging due to the added structural complexities in the inputs as well as the more global context to account for in more ways than one ds fusion demonstrates its ability to utilize all the letters of a word to convey semantic features in a creative manner font specific stylization figures below shows results produced by varying the fonts of the input glyphs for style dragon in letter r maintaing the shape of font including stroke thickness slanting and accents attribute stylization the use of style attributes in the input prompts can be an effective means to further fine tune the stylization this website is licensed under a creative commons attribution sharealike 4 0 international license website source code based on the source code of nerfies