Granted Patent B2

US 12,639,366 · App. 18/838,726 · Granted May 26, 2026

Inventors: Isabella Terrazas (Stanford, CA); Ignacio Gatti (Buenos Aires, AR); John Orozco (Buenos Aires, AR); Rohan Kothakupu (Chicago, IL); Cameron Turner (Palo Alto, CA)

Assignee: Mars, Incorporated

G06F16/538G06F16/532G06F16/583G06V10/40

Abstract

Claims (50)

1 . A computer-implemented method for analyzing a pet sketch to determine one or more matching pet images, the method comprising:

capturing a unique code via a user device having one or more processors;

in response to the capturing, displaying, by the one or more processors, a prompt on the user device for a user to upload or draw a pet sketch;

receiving, by the one or more processors, a digital image of the pet sketch drawn by the user and user location data corresponding to the user device;

utilizing, by the one or more processors, one or more machine-learning models to remove a background of the digital image and analyze the digital image of the pet sketch to determine one or more characteristics of the digital image of the pet sketch;

predicting, by the one or more processors, via the one or more machine-learning models, a pet breed based on the one or more characteristics;

sending, by the one or more processors, an API request to at least one external system, the API request including the one or more characteristics of the digital image of the pet sketch, the user location data, and a characteristics threshold corresponding to a minimum characteristic matching amount;

in response to sending the API request, receiving, by the one or more processors, one or more matching pet images and corresponding data from the at least one external system, the one or more matching pet images matching the pet sketch within the minimum characteristic matching amount defined by the characteristics threshold; and

displaying, by the one or more processors, the predicted pet breed, the one or more matching pet images, and the corresponding data from the at least one external system on the user device.

2 . The computer-implemented method of claim 1 , the method further comprising:

searching, by the one or more processors, one or more databases to find the one or more matching pet images that correspond to the pet sketch, the searching based on the one or more characteristics.

3 . The computer-implemented method of claim 1 , wherein the pet sketch is a user drawing or a pet photograph.

4 . The computer-implemented method of claim 1 , wherein the one or more machine-learning models were trained based on one or more datasets of one or more pet sketches, one or more corresponding characteristics, and one or more corresponding breeds.

5 . The computer-implemented method of claim 1 , wherein the at least one external system includes at least one pet adoption service.

6 . The computer-implemented method of claim 1 , wherein the one or more characteristics include at least one of: a color, a hair type, an ear length, a limb length to body length ratio, tail length and/or shape, a body shape, a head shape, a snout length, a color pattern, an ear shape and/or size, or snout dimensions and/or shape.

7 . The computer-implemented method of claim 1 , wherein the one or more matching pet images include a closest matching pet image determined by the at least one external system based on the user location data, the closest matching pet image corresponding to a matching pet located nearest the user device.

8 . The computer-implemented method of claim 1 , wherein the one or more matching pet images include one or more closest matching pet images determined by the at least one external system based on the user location data, the one or more closest matching pet images corresponding to one or more matching pets located within a proximity of a user location.

9 . The computer-implemented method of claim 1 , wherein the API request specifies only searching for the one or more matching pet images corresponding to one or more pets available for adoption or purchase.

10 . The computer-implemented method of claim 9 , the method further comprising:

displaying, by the one or more processors, at least one link to the at least one external system, the at least one link corresponding to the one or more matching pet images.

11 . The computer-implemented method of claim 1 , the method further comprising:

storing, by the one or more processors, the pet sketch and the one or more matching pet images in one or more databases.

12 . A computer system for analyzing a pet sketch to determine one or more matching pet images, the computer system comprising:

at least one memory storing instructions; and

at least one processor configured to execute the instructions to perform operations comprising:

capturing a unique code via a user device having one or more processors;

in response to the capturing, displaying a prompt on the user device for a user to upload or draw a pet sketch;

receiving a digital image of the pet sketch drawn by the user and user location data corresponding to the user device;

utilizing one or more machine-learning models to remove a background of the digital image and analyze the digital image of the pet sketch to determine one or more characteristics of the digital image of the pet sketch;

predicting, via the one or more machine-learning models, a pet breed based on the one or more characteristics;

sending an API request to at least one external system, the API request including the one or more characteristics of the digital image of the pet sketch, the user location data, and a characteristics threshold corresponding to a minimum characteristic matching amount;

in response to sending the API request, receiving one or more matching pet images and corresponding data from the at least one external system, the one or more matching pet images matching the pet sketch within the minimum characteristic matching amount defined by the characteristics threshold; and

displaying the predicted pet breed, the one or more matching pet images, and the corresponding data from the at least one external system on the user device.

13 . The computer system of claim 12 , wherein the one or more matching pet images include a closest matching pet image determined by the at least one external system based on the user location data, the closest matching pet image corresponding to a matching pet located nearest the user device.

14 . The computer system of claim 12 , wherein the one or more matching pet images include one or more closest matching pet images determined by the at least one external system based on the user location data, the one or more closest matching pet images corresponding to one or more matching pets located within a proximity of a user location.

15 . The computer system of claim 12 , wherein the API request specifies only searching for the one or more matching pet images corresponding to one or more pets available for adoption or purchase.

16 . The computer system of claim 15 , the operations further comprising:

displaying at least one link to the at least one external system, the at least one link corresponding to the one or more matching pet images.

17 . A non-transitory computer-readable medium storing instructions that, when executed by a processor, cause the processor to perform operations for analyzing a pet sketch to determine one or more matching pet images, the operations comprising:

capturing a unique code via a user device having one or more processors;

in response to the capturing, displaying a prompt on the user device for a user to upload or draw a pet sketch;

receiving a digital image of the pet sketch drawn by the user and user location data corresponding to the user device;

utilizing one or more machine-learning models to remove a background of the digital image and analyze the digital image of the pet sketch to determine one or more characteristics of the digital image of the pet sketch;

predicting, via the one or more machine-learning models, a pet breed based on the one or more characteristics;

sending an API request to at least one external system, the API request including the one or more characteristics of the digital image of the pet sketch, the user location data, and a characteristics threshold corresponding to a minimum characteristic matching amount;

in response to sending the API request, receiving one or more matching pet images and corresponding data from the at least one external system, the one or more matching pet images matching the pet sketch within the minimum characteristic matching amount defined by the characteristics threshold; and

displaying the predicted pet breed, the one or more matching pet images, and the corresponding data from the at least one external system on the user device.

18 . The non-transitory computer-readable medium of claim 17 , wherein the pet sketch is a user drawing or a pet photograph.

19 . The non-transitory computer-readable medium of claim 17 , wherein the one or more machine-learning models were trained based on one or more datasets of one or more pet sketches, one or more corresponding characteristics, and one or more corresponding breeds.

20 . The non-transitory computer-readable medium of claim 17 , wherein the one or more characteristics include at least one of: a color, a hair type, an ear length, a limb length to body length ratio, tail length and/or shape, a body shape, a head shape, a snout length, a color pattern, an ear shape and/or size, or snout dimensions and/or shape.

Assignments (1)

Continuity (2)

References Cited (17)