Granted Patent B2

US 12,080,046 · App. 18/414,148 · Granted Sep 3, 2024

Inventors: Elham Saraee (Jamaica Plain, MA); Zachary Halloran (Braintree, MA); Jehan Hamedi (South Boston, MA)

Assignee: VIZIT LABS, INC.

G06V10/761G06F16/438G06N3/045G06V10/40G06V10/82

Abstract

Claims (103)

1. A method, comprising:

receiving, by one or more processors, a pointer to a plurality of images of a web page, the web page configured to display the plurality of images in a first arrangement;

retrieving, by the one or more processors using the pointer, the plurality of images of the web page;

executing, by the one or more processors, at least one machine learning model using the plurality of images from the web page as input to generate at least one image performance score for each image of the plurality of images, the at least one machine learning model trained based on a training set of images labeled based at least on interaction data corresponding to images of the training set of images; and

rearranging, by the one or more processors, the plurality of images on the web page to a second arrangement according to the at least one image performance score generated for each image of the plurality of images of the web page.

2. The method of claim 1 , wherein receiving the pointer comprises:

receiving, by the one or more processors, a data file containing a plurality of URLs from a computing device of an entity hosting the web page;

receiving, by the one or more processors, a page identification of the web page; or

receiving, by the one or more processors, a product identification corresponding to a record in a database, the record containing the plurality of images.

3. The method of claim 1 , wherein receiving the pointer comprises receiving, by the one or more processors, a data file containing a plurality of image files containing images on the web page from a computing device of an entity hosting the web page.

4. The method of claim 1 , further comprising:

determining, by the one or more processors, the second arrangement based on the at least one image performance score generated for each image of the plurality of images of the web page by:

comparing, by the one or more processors, each of the at least one image performance score generated for each image of the plurality of images of the web page;

ranking, by the one or more processors, each of the plurality of images of the web page based on the comparison; and

determining, by the one or more processors, the second arrangement based on the ranking of each of the plurality of images.

5. The method of claim 1 , wherein executing the at least one machine learning model using the plurality of images of the web page as input comprises executing, by the one or more processors, a plurality of machine learning models to generate a plurality of image performance scores for each image of the plurality of images, each machine learning model trained to output images for a different target audience corresponding to the machine learning model.

6. The method of claim 5 , further comprising:

determining, by the one or more processors, the second arrangement based on the plurality of image performance scores generated for each of the plurality of images by:

comparing, by the one or more processors, the plurality of image performance scores for each image of the web page with the plurality of image performance scores for each other image of the web page;

ranking, by the one or more processors, each of the plurality of images of the web page based on the comparison; and

determining, by the one or more processors, the second arrangement based on the ranking of each of the plurality of images.

7. The method of claim 1 , further comprising:

provisioning, by the one or more processors, executable code to a computing device hosting the web page, the executable code configured to change arrangement of images of the web page,

wherein rearranging the plurality of images on the web page comprises transmitting, by the one or more processors, the second arrangement of the web page to the computing device, receipt of the second arrangement causing the provisioned executable code stored in memory of the computing device to rearrange the plurality of images on the web page according to the second arrangement.

8. The method of claim 7 , wherein the computing device rearranges the plurality of images of the web page according to the second arrangement for the web page by moving one or more placeholders for loading the plurality of images on the web page according to the second arrangement or by moving the plurality of images as embedded on the web page.

9. The method of claim 1 , further comprising:

subsequent to rearranging the plurality of images of the web page to the second arrangement, receiving, by the one or more processors, upload counter data indicating a number of uploads of a second web page corresponding to the web page; and

generating, by the one or more processors from the upload counter data, a record indicating the number of uploads of the second web page and an identification of the second arrangement of the first plurality of images for the web page.

10. The method of claim 1 , further comprising:

receiving, by the one or more processors, first interaction data of the web page with the plurality of images in the first arrangement;

subsequent to rearranging the plurality of images on the web page to the second arrangement, receiving, by the one or more processors, second interaction data of the web page with the plurality of images in the second arrangement; and

generating, by the one or more processors, a record indicating a difference between the first interaction data and the second interaction data.

11. The method of claim 1 , comprising:

subsequent to rearranging the plurality of images on the web page to the second arrangement, receiving, by the one or more processors, interaction data of the web page with the plurality of images in the second arrangement;

comparing, by the one or more processors, the interaction data of the web page in the second arrangement to a threshold; and

responsive to determining the interaction data of the web page in the second arrangement is less than the threshold, rearranging, by the one or more processors, the plurality of images of the web page according to a third arrangement.

12. The method of claim 1 , further comprising:

receiving, by the one or more processors from a pixel placed on the web page, interaction data of the web page subsequent to rearranging the plurality of images;

determining, by the one or more processors, the interaction data satisfies a condition; and

rearranging, by the one or more processors, the plurality of images on the web page according to a third arrangement responsive to determining the interaction data satisfies the condition.

13. The method of claim 12 , wherein the at least one machine learning model is a first at least one machine learning model and the at least one image performance score is a first at least one image performance score, the method further comprising:

executing, by the one or more processors, a second at least one machine learning model using the plurality of images from the web page as input to generate a second at least one image performance score for each image of the first plurality of images,

wherein rearranging the first plurality of images on the web page according to the third arrangement comprises rearranging, by the one or more processors, the first plurality of images according to the second at least one image performance score for each image of the first plurality of images.

14. The method of claim 1 , further comprising:

determining, by the one or more processors, a web page score for the web page as a function of the at least one image performance score for each image of the plurality of images; and

transmitting, by the one or more processors, the web page score to an entity hosting the web page.

15. The method of claim 1 , further comprising:

executing, by the one or more processors, the at least one machine learning model using a set of images from each of a plurality of web pages to generate an image performance score for each image of each set of images;

determining, by the one or more processors, a web page score for each of the plurality of web pages as a function of the image performance scores generated for each image of the set of images of the web page;

ranking, by the one or more processors, each of the plurality of web pages based on the determined web page scores; and

generating, by the one or more processors, a record indicating the rankings of the plurality of web pages.

16. A system, comprising:

one or more processors coupled to memory and configured to:

receive a pointer to a plurality of images of a web page, the web page configured to display the plurality of images in a first arrangement;

retrieve, using the pointer, the plurality of images of the web page;

execute at least one machine learning model using the plurality of images from the web page as input to generate at least one image performance score for each image of the plurality of images, the at least one machine learning model trained based on a training set of images labeled based at least on interaction data corresponding to images of the training set of images; and

rearrange the plurality of images on the web page to a second arrangement according to the at least one image performance score generated for each image of the plurality of images of the web page.

17. The system of claim 16 , wherein one or more processors are configured to receive the pointer by:

receiving a data file containing a plurality of URLs from a computing device of an entity hosting the web page;

receiving a page identification of the web page; or

receiving a product identification corresponding to a record in a database, the record containing the plurality of images.

18. The system of claim 16 , wherein the one or more processors are further configured to:

determine the second arrangement based on the at least one image performance score generated for each image of the plurality of images of the web page by:

comparing each of the at least one image performance scores generated for each images of the plurality of images the web page;

ranking each of the plurality of images of the web page based on the comparison; and

determining the second arrangement based on the ranking of each of the plurality of images.

19. A non-transitory computer-readable storage medium having instructions embodied thereon, the instructions being executable by one or more processors to perform a method, the method comprising:

receiving a pointer to a plurality of images of a web page, the web page configured to display the plurality of images in a first arrangement;

retrieving, using the pointer, the plurality of images of the web page;

executing at least one machine learning model using the plurality of images from the web page as input to generate at least one image performance score for each image of the plurality of images, the at least one machine learning model trained based on a training set of images labeled based at least on interaction data corresponding to images of the training set of images; and

rearranging the plurality of images on the web page to a second arrangement according to the at least one image performance score generated for each image of the plurality of images of the web page.

20. The non-transitory computer-readable storage medium of claim 19 , wherein the method further comprises:

determining the second arrangement based on the at least one image performance score generated for each image of the plurality of images of the web page by:

comparing each of the at least one image performance scores of the web page;

ranking each of the plurality of images of the web page based on the comparison; and

determining the second arrangement based on the ranking of each of the plurality of images.

21. A method, comprising:

receiving, by one or more processors, an identification of a first product in a request to generate image recommendations for the first product;

retrieving, by the one or more processors, a plurality of images from a record in a database based on the record containing the identification of the first product;

executing, by the one or more processors, at least one machine learning model using the plurality of images depicting the first product as input to generate at least one image performance score for each image of the plurality of images, the at least one machine learning model trained based on a training set of images labeled based at least on interaction data corresponding to images of the training set of images;

ranking, by the one or more processors, each of the plurality of images depicting the first product based on the at least one image performance score for each of the plurality of images to generate a ranked list of the plurality of images;

based on the rankings of the plurality of images, determining, by the one or more processors, a subset of images for each of a plurality of servers according to image display policies of the plurality of servers, each of the plurality of servers corresponding to a different website; and

transmitting, by the one or more processors to each of the plurality of servers, the subset of images determined for the server, the server displaying the subset of images on a web page corresponding to the first product.

22. The method of claim 21 , wherein executing the at least one machine learning model using the plurality of images depicting the first product as input comprises executing, by the one or more processors, a plurality of machine learning models to generate a plurality of image performance scores for each image of the plurality of images, each machine learning model trained to output images for a different target audience corresponding to the machine learning model.

23. The method of claim 22 , wherein ranking each of the plurality of images comprises:

determining, by the one or more processors, an aggregate image performance score for each of the plurality of images as a function of the plurality of image performance scores; and

ranking, by the one or more processors, each of the plurality of images based on the aggregate image performance score for each of the plurality of images.

24. The method of claim 22 , wherein ranking each of the plurality of images comprises:

ranking, by the one or more processors, each of the plurality of images for each of the target audiences based on image performance scores generated for the plurality of images.

25. The method of claim 22 , wherein a server of the plurality of servers corresponds to a different image display policy indicating a target audience of the website corresponding to the server;

wherein ranking each of the plurality of images comprises:

ranking, by the one or more processors, each of the plurality of images based on image performance scores generated for the plurality of images for the target audience; and

wherein transmitting the ranked list of the plurality of images to the server comprises:

transmitting the rankings of the plurality of images determined based on the image performance scores generated for the plurality of images for the target audience.

26. The method of claim 21 , wherein a server of the plurality of servers displays the plurality of images based on an order of the plurality of images in the ranked list.

27. The method of claim 21 , further comprising:

obtaining, by the one or more processors, a second image depicting the first product;

executing, by the one or more processors, the at least one machine learning model using the second image as input to determine a second image performance score for the second image;

comparing, by the one or more processors, the second image performance score with the at least one image performance score for each of the plurality of images; and

ranking, by the one or more processors, the second image with the plurality of images based on the comparison.

28. The method of claim 21 , wherein each of the plurality of images depicting the first product correspond to a stock keeping unit (SKU) in the database.

29. The method of claim 21 , wherein determining the subset of images for the server comprises:

identifying, by the one or more processors, a number of images to recommend to the server according to an image display policy of the server.

Assignments (1)

Continuity (9)

Cited By (14)

US 12,322,402 US 12,423,388 US 12,505,203 US 12,505,653 US 12,632,514 US 12,632,515 US 12,632,516 US 12,632,517 US 12,632,518 US 12,633,275 US 12,633,276 US 12,639,405 US 12,645,766 US 12,682,636