Granted Patent B2

US 12,646,100 · App. 19/191,539 · Granted Jun 2, 2026

Inventor: Neha Konakalla (Cupertino, CA)

Assignee: Pyxer Inc.

G06Q30/0631G06Q30/0629G06T3/18G06T5/77G06T19/006

Abstract

Claims (64)

1 . A computer-implemented method for providing generative clothing recommendations, via one or more processors, configured to perform the operations of:

receiving one or more images, via a user interface of an application executed on a client device, the one or more images depicting one or more clothing items;

identifying and segmenting regions of clothing components from the received one or more images;

determining clothing characteristics of the identified clothing components, the clothing characteristics comprising a clothing type and a color, wherein the clothing characteristics are associated with a user;

utilizing a conversational artificial intelligence (AI) model to generate one or more conversational interactions with the user, via a user interface of the application, wherein output generated from the conversational AI model is displayed via the user interface;

generating by a visual AI model, one or more new images depicting one or more recommended clothing items for the user, wherein the recommended clothing items are based on the determined clothing characteristics and the one or more conversational interactions, wherein the generated one or more new images include a depiction of the user that is based on the one or more images that were received by the application, wherein the one or more recommended clothing items are virtually generated pieces of clothing that are placed upon at least a subset of the segmented regions of the received one or more images;

providing for display, via the user interface of the application, the generated one or more new images;

receiving a selection of an image, via the user interface, of the generated one or more new images;

based on the selected generated image, determining one or more inventory items; and

displaying at least one of the determined inventory items, via the user interface.

2 . The method of claim 1 , further comprising:

determining a set of clothing preferences from the one or more conversational interactions with the user, wherein the generated one or more new images are also based on the determined set of clothing preferences.

3 . The method of claim 1 , further comprising:

iteratively training a machine learning model to identify the clothing characteristics based on the received one or more input images, wherein the clothing characteristics comprise clothing patterns and/or clothing components;

inputting the received one or more images into the trained machine learning model; and

recommending a set of clothing items based on the identified clothing characteristics.

4 . The method of claim 1 , further comprising:

providing for display, via the user interface of the application, new recommended clothing items based on the conversational interactions with the user, wherein the set of clothing preferences of the clothing items are based at least in part on the one or more additional conversational interactions and based on the one or more recommended clothing items.

5 . The method of claim 1 , further comprising:

providing for display, via the user interface, a dialogue with an explanation for the recommended one or more clothing items as to why a particular piece of clothing was recommended.

6 . The method of claim 1 , further comprising:

wherein the user interfaces provides interface controls to receive a like or dislike of the recommended one or more clothing items; and

using a received like or dislike input via the user interface of the application to recalibrate subsequent recommendations of clothing items.

7 . The method of claim 1 , further comprising:

segmenting each of the received one or more images into distinct regions corresponding to different clothing items; and

extracting clothing components and isolating the clothing components from a background image.

8 . The method of claim 1 , further comprising:

extracting clothing component characteristics from the received one or more images; and

constructing a profile of fashion preferences that provides a basis for personalized clothing recommendation.

9 . The method of claim 1 , wherein the recommended clothing items are also based on weather or environmental conditions.

10 . The method of claim 1 , further comprising:

receiving a set of clothing purchase data, wherein the recommended clothing items are based on the received set of clothing purchase data.

11 . The method of claim 1 , wherein placement of the virtually generated pieces of clothing comprises: overlaying and/or in-painting the virtually generated pieces of clothing about the segmented regions of the received one or more images.

12 . A system comprising one or more processors configured to perform the operations of:

receiving one or more images, via a user interface of an application executed on a client device, the one or more images depicting one or more clothing items;

identifying and segmenting regions of clothing components from the received one or more images;

determining clothing characteristics of the identified clothing components, the clothing characteristics comprising a clothing type and a color, wherein the clothing characteristics are associated with a user;

utilizing a conversational artificial intelligence (AI) model to generate one or more conversational interactions with the user, via a user interface of the application, wherein output generated from the conversational AI model is displayed via the user interface;

generating by a visual AI model, one or more new images depicting one or more recommended clothing items for the user, wherein the recommended clothing items are based on the determined clothing characteristics and the one or more conversational interactions, wherein the generated one or more new images include a depiction of the user that is based on the one or more images that were received by the application, wherein the one or more recommended clothing items are virtually generated pieces of clothing that are placed upon at least a subset of the segmented regions of the received one or more images;

providing for display, via the user interface of the application, the generated one or more new images;

performing a similarity match of the recommended clothing items and one or more inventory items; and

displaying at least one of the inventory items as an output of the similarity match.

13 . The system of claim 12 , further comprising the operations of:

determining a set of clothing preferences from the one or more conversational interactions with the user, wherein the generated one or more new images are also based on the determined set of clothing preferences.

14 . The system of claim 12 , further comprising the operations of:

iteratively training a machine learning model to identify the clothing characteristics based on the received one or more images, wherein the clothing characteristics comprise clothing patterns and/or clothing components;

inputting the received one or more images into the trained machine learning model; and

recommending a set of clothing items based on the identified clothing characteristics.

15 . The system of claim 12 , further comprising the operations of:

providing for display, via the user interface of the application, new recommended clothing items based on the conversational interactions with the user, wherein the set of clothing preferences of the clothing items are based at least in part on one or more additional conversational interactions and based on the one or more recommended clothing items.

16 . The system of claim 12 , further comprising the operations of:

providing for display, via the user interface, a dialogue with an explanation for the recommended one or more clothing items as to why a particular piece of clothing was recommended.

17 . The system of claim 12 , further comprising the operations of:

wherein the user interface provides interface controls to receive a like or dislike of the recommended one or more clothing items; and

using a received like or dislike input via the user interface of the application to recalibrate subsequent recommendations of clothing items.

18 . The system of claim 12 , further comprising the operations of:

segmenting each of the received one or more images into distinct regions corresponding to different clothing items; and

extracting of clothing components and isolating the clothing components from a background image.

19 . The system of claim 12 , further comprising the operations of:

extracting clothing component characteristics from the received one or more images; and

constructing a profile of fashion preferences that provides a basis for personalized clothing recommendations.

20 . The system of claim 12 , wherein the recommended clothing items are also based on weather or environmental conditions.

21 . The system of claim 12 , further comprising:

receiving a set of clothing purchase data, wherein the recommended clothing items are based on the received set of clothing purchase data.

Continuity (3)

References Cited (12)