TextStyleBrush: Transfer of Text Aesthetics From a Single Example

Praveen Krishnan, Rama Kovvuri, Guan Pang, Boris Vassilev, Tal Hassner

פרסום מחקרי: פרסום בכתב עתמאמרביקורת עמיתים


We present a novel approach for disentangling the content of a text image from all aspects of its appearance. The appearance representation we derive can then be applied to new content, for one-shot transfer of the source style to new content. We learn this disentanglement in a self-supervised manner. Our method processes entire word boxes, without requiring segmentation of text from background, per-character processing, or making assumptions on string lengths. We show results in different text domains which were previously handled by specialized methods, e.g., scene text, handwritten text. To these ends, we make a number of technical contributions: (1) We disentangle the style and content of a textual image into a non-parametric, fixed-dimensional vector. (2) We propose a novel approach inspired by StyleGAN but conditioned over the example style at different resolution and content. (3) We present novel self-supervised training criteria which preserve both source style and target content using a pre-trained font classifier and text recognizer. Finally, (4) we also introduce Imgur5K, a new challenging dataset for handwritten word images. We offer numerous qualitative photo-realistic results of our method. We further show that our method surpasses previous work in quantitative tests on scene text and handwriting datasets, as well as in a user study.

שפה מקוריתאנגלית
עמודים (מ-עד)9122-9134
מספר עמודים13
כתב עתIEEE Transactions on Pattern Analysis and Machine Intelligence
מספר גיליון7
מזהי עצם דיגיטלי (DOIs)
סטטוס פרסוםפורסם - 1 יולי 2023

הערה ביבליוגרפית

Publisher Copyright:
© 1979-2012 IEEE.

טביעת אצבע

להלן מוצגים תחומי המחקר של הפרסום 'TextStyleBrush: Transfer of Text Aesthetics From a Single Example'. יחד הם יוצרים טביעת אצבע ייחודית.

פורמט ציטוט ביבליוגרפי