Content aware font recommendation
AssigneeAdobe Inc.
InventorsNeel Kadia, Shikhar Garg, Saikat Chakrabarty
Life of this patent
Estimating the term…
Abstract
Font recommendation systems and techniques are provided for determining one or more font recommendations. Content aware font recommendations can be determined for an electronic document based on information from the design context of the electronic document. For example, to generate recommended fonts for an electronic document, the font recommendation systems and techniques can obtain one or images of the electronic document and/or text from the electronic document as input, and can process the one or more images and/or the text to determine recommended font tags. In some cases, recommended fonts can be determined based on user search queries, such as when a user searches for a particular font.
Claims16
A method of determining one or more font recommendations, comprising:
- obtaining an electronic document comprising one or more images and text;
- processing the one or more images of the electronic document using a plurality of neural networks to determine one or more image tags from the electronic document, each neural network of the plurality of neural networks determining a different category of a set of categories of image tags, wherein the set of categories include one or more of a user group, genre, media, or mood of the one or more images, the image tags representing classes associated with the one or more images;
- determining one or more font tags based on one or more tag scores computed between the one or more font tags and the one or more image tags from the electronic document using a plurality of image tag to font tag maps, wherein the one or more font tags include one or more keywords used to search for fonts;
- processing the text of the electronic document using a text tag generation engine to generate a set of words of interest to be used as text font tags for the electronic document
- augmenting the one or more font tags to include the text font tags;
- and
- determining one or more fonts to recommend for the electronic document using the one or more font tags.
A system for determining one or more font recommendations, comprising:
- one or more processors;
- and
- memory accessible to the one or more processors, the memory storing instructions, which upon execution by the one or more processors, cause the one or more processors to:
- obtain an electronic document comprising one or more images and text;
- process the one or more images of the electronic document using a plurality of neural networks to determine one or more image tags from the electronic document, each neural network of the plurality of neural networks determining a different category of a set of categories of image tags, wherein the set of categories include one or more of a user group, genre, media, or mood of the one or more images, the image tags representing classes associated with the one or more images;
- determine one or more font tags based on one or more tag scores computed between the one or more font tags and the one or more image tags from the electronic document using a plurality of image tag to font tag maps, wherein the one or more font tags include one or more keywords used to search for fonts;
- process the text of the electronic document using a text tag generation engine to generate a set of words of interest to be used as text font tags for the electronic document;
- augment the one or more font tags to include the text font tags;
- and
- determine one or more fonts to recommend for the electronic document using the one or more font tags.
A non-transitory computer-readable medium having stored thereon instructions that, when executed by one or more processors, cause the one or more processors to:
- obtain an electronic document comprising one or more images and text;
- process the one or more images of the electronic document using a plurality of neural networks to determine one or more image tags from the electronic document, each neural network of the plurality of neural networks determining a different category of a set of categories of image tags, wherein the set of categories include one or more of a user group, genre, media, or mood of the one or more images, the image tags representing classes associated with the one or more images;
- determine one or more font tags based on one or more tag scores computed between the one or more font tags and the one or more image tags from the electronic document using a plurality of image tag to font tag maps, wherein the one or more font tags include one or more keywords used to search for fonts;
- process the text of the electronic document using a text tag generation engine to generate a set of words of interest to be used as text font tags for the electronic document
- augment the one or more font tags to include the text font tags;
- and
- determine one or more fonts to recommend for the electronic document using the one or more font tags.
Dependencies are inferred from the claim text.