US 12,260,480 B2
Machine learning-based layout generation
Sukriti Verma, Pittsburgh, PA (US); Venkata naveen kumar Yadav Marri, San Jose, CA (US); Ritiz Tambi, San Francisco, CA (US); Pranav Vineet Aggarwal, Santa Clara, CA (US); Peter O'Donovan, Seattle, WA (US); Midhun Harikumar, Sunnyvale, CA (US); and Ajinkya Kale, San Jose, CA (US)
Assigned to Adobe Inc., San Jose, CA (US)
Filed by Adobe Inc., San Jose, CA (US)
Filed on Mar. 6, 2023, as Appl. No. 18/178,791.
Prior Publication US 2024/0303881 A1, Sep. 12, 2024
Int. Cl. G06T 11/60 (2006.01); G06F 3/0482 (2013.01)
CPC G06T 11/60 (2013.01) [G06F 3/0482 (2013.01); G06T 2200/24 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A method comprising:
receiving, via a user interface, a user selection of a set of design elements for performing generative layout recommendation;
generating a set of tokens representing design elements of the set of design elements;
generating, using a first machine learning model, a token embedding and a position embedding for each token of the set of tokens;
generating, by a second machine learning model, a sequence of tokens corresponding to positions and orientations of the set of design elements using the token embedding and the position embedding for each token of the set of tokens;
rendering a layout according to the sequence of tokens;
receiving a selection of a similarity metric that represents a requested similarity or difference between a set of recommended layouts and the set of design elements;
creating one or more recommended layouts for the set of recommended layouts, wherein each layout in the one or more recommended layouts is created using the similarity metric; and
outputting the set of recommended layouts to a user via the user interface, wherein the set of recommended layouts includes one or more rendered layouts and wherein each layout of the one or more rendered layouts includes an arrangement of the user selected set of design elements in a graphical output visualization.