Image evaluation

• US 10,176,406 B2
• Filed: 05/10/2016
• Issued: 01/08/2019
• Est. Priority Date: 04/04/2014
• Status: Active Grant

• Current Assignee: eBay Inc.
• Original Assignee: eBay Inc.
• Inventors: Di, Wei, Jagadeesh, Vignesh, Piramuthu, Robinson, Churchill, Elizabeth, Bhardwaj, Anurag
• Primary Examiner(s): Patel, Jayesh A
• Application Number:
  US15/151,435
• Publication Number:
  US 20160253582A1
• Time in Patent Office: 973 Days
• Field of Search: None
• CPC Class Codes: G06F 16/58 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/5854 using shape and object relationship
• CPC Class Codes: G06F 16/60 of audio data
• CPC Class Codes: G06F 16/63 Querying
• CPC Class Codes: G06F 18/24 Classification techniques
• CPC Class Codes: G06F 18/2415 based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]
• CPC Class Codes: G06Q 30/0613 Third-party assisted
• CPC Class Codes: G06V 30/192 using simultaneous comparisons or correlations of the image signals with a plurality of references
• CPC Class Codes: G06V 30/198 the selection of the next reference depending on the result of the preceding comparison

Abstract

A machine may be configured to perform image evaluation of images depicting items for online publishing. For example, the machine performing a user behavior analysis based on data pertaining to interactions by a plurality of users with a plurality of images pertaining to a particular type of item. The machine determines, based on the user behavior analysis, that a presentation type associated with one or more images of the plurality of images corresponds to a user behavior in relation to the one or more images. The machine determines that an item included in a received image is of the particular type of item. The machine generates an output for display in a client device. The output includes a reference to the received image and a recommendation of the presentation type for the item included in the received image, for publication by a web server of a publication system.

Claims

1. A system comprising:

a non-transitory machine-readable medium for storing instructions that, when executed by one or more hardware processors of a machine, cause the one or more hardware processors to perform operations comprising;

accessing user behavior data pertaining to interactions by a plurality of users with a plurality of test images pertaining to a particular type of item;

performing an analysis of the interactions;

generating results of the analysis of the interactions, the results indicating a recommended presentation type for the particular type of item;

storing the results of the analysis of the interactions in a database;

receiving an image of an item from a client device;

extracting one or more visual features from the received image;

determining that the item included in the received image is of the particular type of item based on the extracted one or more visual features; and

generating, based on the determining that the item included in the received image is of the particular type of item, an output for display in the client device, the output including a reference to the received image and the recommended presentation type for the item included in the received image of the item based on the results of the analysis of the interactions, wherein the presentation type corresponds to an attribute-value pair of one or more attribute-value pairs associated with one or more images of the plurality of test images corresponding to the user behavior, and wherein the operations further comprise;

determining, for the received image, a likelihood of a user engaging in the user behavior in relation to the received image based on one of the one or more attribute-value pairs associated with the received image.

2. The system of claim 1, wherein the operations further comprise:

determining a likelihood of a user engaging in the user behavior based on a publication, by the web server of the publishing system, of an image of the particular type of item using the presentation type,

wherein the output further includes a reference to the likelihood of a user engaging in the user behavior.

3. The system of claim 1, wherein the operations further comprise:

performing an evaluation of the received image based on comparing the received image and one or more other images.

4. The system of claim 3, wherein the received image and the one or more other images are received from a client device associated with a particular user.

5. The system of claim 3, wherein the received image is received from a first client device associated with a first user and the one or more other images are received from a second client device associated with one or more other users.

6. The system of claim 3, wherein the output includes a result of the comparing of the received image and the one or more other images.

7. The system of claim 1, wherein the operations further comprise:

performing an evaluation of the received image based on an image analysis of the received image, the image analysis including identifying one or more image attributes and one or more values that correspond to the one or more image attributes, the one or more image attributes comprising at least one of an image quality attribute and a display type attribute, the display type attribute indicating a visual context in which the item is displayed.

8. The system of claim 7, wherein the determining that the item included in the received image is of the particular type of item includes classifying the received image into a category based on the one or more image attributes and the one or more values that correspond to the one or more image attributes.

9. The system of claim 7, wherein the generating of the output is further based on a result of the evaluation of the received image.

10. The system of claim 7, wherein the output includes a further recommendation to provide a further image with a different value corresponding to an attribute of the one or more attributes.

11. The system of claim 1, wherein the output includes a further recommendation of an action pertaining to the received image to increase the likelihood of the user engaging in the user behavior.

12. A method comprising:

accessing user behavior data pertaining to interactions by a plurality of users with a plurality of test images pertaining to a particular type of item;

performing an analysis of the interactions;

generating results of the analysis of the interactions, the results indicating a recommended presentation type for the particular type of item;

storing the results of the analysis of the interactions in a database;

receiving an image of an item from a client device;

extracting one or more visual features from the received image;

determining, using one or more hardware processors, that the item included in the received image is of the particular type of item based on the extracted one or more visual features; and

generating, based on the determining that the item included in the received image is of the particular type of item, an output for display in the client device, the output including a reference to the received image and the recommended presentation type for the item included in the received image of the item based on the results of the analysis of the interactions wherein the presentation type corresponds to an attribute-value pair of one or more attribute-value pairs associated with one or more images of the plurality of test images corresponding to the user behavior, the method further comprising;

determining, for the received image, a likelihood of a user engaging in the user behavior hi relation to the received image based on one of the one or more attribute-value pairs associated with the received image.

13. The method of claim 12, further comprising:

determining a likelihood of a user engaging in the user behavior based on a publication, by the web server of the publishing system, of an image of the particular type of item using the presentation type,

wherein the output further includes a reference to the likelihood of a user engaging in the user behavior.

14. The method of claim 12, further comprising:

performing an evaluation of the received image based on comparing the received image and one or more other images.

15. The method of claim 12, further comprising:

performing an evaluation of the received image based on an image analysis of the received image, the image analysis including identifying one or more image attributes and one or more values that correspond to the one or more image attributes, the one or more image attributes comprising at least one of an image quality attribute and a display type attribute, the display type attribute indicating a visual context in which the item is displayed.

16. A non-transitory computer-readable storage medium storing instructions which, when executed by one or more hardware processors of a machine, cause the machine to perform operations comprising:

performing a user behavior analysis based on historical data pertaining to interactions by a plurality of users with a plurality of images pertaining to a particular type of item;

determining, based on the user behavior analysis, that a presentation type associated with one or more images of the plurality of images corresponds to a user behavior in relation to the one or more images;

receiving an image of an item from a client device;

extracting one or more visual features from the received image;

determining that the item included in the received image is of the particular type of item based on the extracted one or more visual features; and

generating, based on the determining that the item included in the received image is of the particular type of item, an output for display in the client device, the output including a reference to the received image and the recommended presentation type for the item included in the received image of the item based on the results of the analysis of the interactions, wherein the presentation type corresponds to an attribute-value pair of one or more attribute-value pairs associated with the one or more images corresponding to the user behavior, and wherein the operations further comprising;

determining, for the received image, a likelihood of a user engaging in the user behavior in relation to the received image based on one of the one or more attribute-value pairs associated with the received image.