Automatically classifying and presenting digital fonts

Patent metadata

Field	Value
Publication number	US-10489489-B2
Application number	US-201615065501-A
Country	US
Kind code	B2
Filing date	Mar 9, 2016
Priority date	Mar 9, 2016
Publication date	Nov 26, 2019
Grant date	Nov 26, 2019

Abstract

Systems and methods are disclosed for classifying digital fonts. In particular, in one or more embodiments, the disclosed systems and methods detect a new digital font, automatically classify the digital font into one or more font classifications, and make the digital font available via a user interface. More particularly, the disclosed systems and methods can conduct searches for the new digital font, identify digital fonts similar to the new digital font, and apply the new digital font to digital text in an electronic document.

First claim

We claim:

1. In a digital medium environment for creating or editing digital documents, a method of classifying and presenting electronic fonts, comprising:
detecting, by a processor, a new digital font on a computing device;
generating, by the processor and a classification neural network, one or more neural network feature vectors corresponding to the new digital font based on font features of the new digital font;
determining, utilizing the classification neural network, a plurality of font classifications based on the one or more neural network feature vectors, the plurality of font classifications comprising a font attribute and a font class;
providing for display, a user interface of an electronic document application on the computing device, wherein the user interface comprises text rendered utilizing an existing font and selectable font search elements, the selectable font search elements comprising a similar digital font filter element; and
in response to receiving user input of a search query via the similar digital font filter element of the user interface of the electronic document application:
identifying a similar digital font based on a distance between the one or more neural network feature vectors determined utilizing the classification neural network and an additional neural network feature vector corresponding to the existing font; and
rendering, by the processor and within a list of fonts of the user interface of the electronic document application, a visual representation of the similar digital font and two or more digital fonts for selection and application to digital text.

2. The method of claim 1 , wherein the selectable font search elements further comprise a font attribute selectable element, a font class selectable element, a server font filter selectable element, and a favorites font selectable element.

3. The method of claim 2 , wherein the similar digital font consists of the new digital font.

4. The method of claim 2 , further comprising receiving a second user input, via the font class selectable element, of one or more font classes from a set of font classes, the set of font classes comprising serif, sans serif, and slab serif; and
conducting a search based on the one or more font classes.

5. The method of claim 2 , further comprising generating the additional neural network feature vector corresponding to the existing font utilizing the classification neural network.

6. The method of claim 1 , wherein the classification neural network comprises a convolutional neural network trained to generate one or more font classifications based on training fonts and training font classifications.

7. The method of claim 2 , further comprising:
receiving a second user input via the favorites font selectable element; and
in response to the second user input, rendering a plurality of fonts based on previous user interaction with the plurality of fonts.

8. The method of claim 2 , further comprising:
in response to receiving a second user input via the font attribute selectable element, providing visual representations of a plurality of font attributes, wherein the visual representations of the plurality of font attributes comprise descriptive terms;
upon selection of the font attribute from the plurality of font attributes, identifying additional digital fonts from a plurality of digital fonts on the computing device and the new digital font that correspond to the font attribute; and
providing a visual representation of the additional digital fonts for display in the list of fonts.

9. A system that allows computing devices to classify and present electronic fonts comprising:
at least one processor; and
at least one non-transitory computer readable storage medium storing instructions that, when executed by the at least one processor, cause the system to:
search a repository of digital fonts for a new digital font;
upon identifying the new digital font in the repository of digital fonts, utilize a classification neural network to determine one or more neural network feature vectors corresponding to the new digital font based on one or more font features of the new digital font;
determine, utilizing the classification neural network, font classifications for the new digital font based on the one or more neural network feature vectors;
provide in a user interface of an electronic document application a list of fonts, text rendered utilizing an existing font and selectable font search elements, the selectable font search elements comprising a similar digital font filter element;
in response to on user selection of the similar digital font filter element of the user interface:
identify a similar digital font based on a distance between the one or more neural network feature vectors determined utilizing the classification neural network and an additional neural network feature vector corresponding to the existing font; and
provide in the user interface of the electronic document application a visual representation of the similar digital font and two or more digital fonts.

10. The system of claim 9 , wherein the selectable font search elements further comprise a font attribute selectable element, a font class selectable element, a server font filter selectable element, and a favorites font selectable element.

11. The system of claim 10 , further comprising instructions that, when executed by the at least one processor, cause the system to:
receive a second user input comprising user interaction via the server font filter selectable element; and
render a set of fonts available via a remote server.

12. The system of claim 9 , wherein the classification neural network comprises a convolutional neural network trained to generate one or more font classifications based on training fonts and training font classifications.

13. The system of claim 9 , further comprising instructions that, when executed by the at least one processor, cause the system to generate the additional neural network feature vector corresponding to the existing font utilizing the classification neural network.

14. The system of claim 9 , further comprising instructions that, when executed by the at least one processor, cause the system to:
detect a plurality of new digital fonts;
rank the plurality of new digital fonts into a priority order based on whether each of the plurality of new digital fonts was created on a client device; and
determine a plurality of font classifications corresponding to each of the new digital fonts by analyzing each new digital font according to the priority order.

15. The system of claim 14 , further comprising instructions that, when executed by the at least one processor, cause the system to rank the plurality of new digital fonts into the priority order based on file size of each of the plurality of new digital fonts.

16. The system of claim 10 , further comprising instructions that, when executed by the at least one processor, cause the system to:
in response to user interaction with the font attribute selectable element, provide visual representations of a plurality of font attributes with the list of fonts, wherein the visual representations of the plurality of font attributes comprise a plurality of descriptive terms, wherein the descriptive terms comprise at least one of: angular, artistic, attention-grabbing, attractive, boring, calm, charming, clumsy, complex, delicate, disorderly, dramatic, formal, fresh, friendly, gentle, graceful, happy, modern, playful, pretentious, or sloppy;
upon selection of at least one font attribute of the font attributes

Assignees

• Adobe Inc

Inventors

• Song Yuyan
• Shaw Seth
• Hertzmann Aaron

Classifications

• G06F16/332 Primary

  Query formulation · CPC title

• G06F40/109 Primary

  Font handling; Temporal or kinetic typography · CPC title

• G06F40/166

  Editing, e.g. inserting or deleting · CPC title

• G06F3/04842

  Selection of displayed objects or displayed text elements (G06F3/0482 takes precedence) · CPC title

• G06N20/00

  Machine learning · CPC title