US 12,266,034 B2
Techniques for abstract image generation from multimodal inputs with content appropriateness considerations
Julia Gong, Cary, NC (US); Houdong Hu, Redmond, WA (US); and William Douglas Guyman, Seattle, WA (US)
Assigned to Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed by Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed on Jul. 30, 2022, as Appl. No. 17/877,935.
Prior Publication US 2024/0037810 A1, Feb. 1, 2024
Int. Cl. G06T 11/00 (2006.01); G06F 16/53 (2019.01); G06F 16/583 (2019.01); G06F 40/20 (2020.01); G06F 40/30 (2020.01); G06T 7/00 (2017.01); G06T 7/90 (2017.01)
CPC G06T 11/001 (2013.01) [G06F 16/53 (2019.01); G06F 16/5838 (2019.01); G06F 40/20 (2020.01); G06F 40/30 (2020.01); G06T 7/0002 (2013.01); G06T 7/90 (2017.01); G06T 2207/10024 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] 23 Claims
OG exemplary drawing
 
1. A data processing system comprising:
a processor; and
a machine-readable medium storing executable instructions that, when executed, cause the processor to perform operations comprising:
receiving a textual input comprising a query for a first image;
analyzing the textual input to determine a predicted color palette associated with a subject matter of the query by:
providing the textual input to a first machine learning model, the first machine learning model trained to analyze the textual input and to output a predicted color palette length, the predicted color palette length predicting a number of colors to represent a subject matter of the query based on a complexity of the query; and
analyzing the textual input with the first machine learning model to obtain the predicted color palette length; and
procedurally generating the first image using the predicted color palette by algorithmically generating the first image using a procedural generation algorithm that places limits on types of content to be included in the first image and a style of the first image.