Finding products that are similar to a product selected from a plurality of products

Patent metadata

Field	Value
Publication number	US-9715510-B2
Application number	US-201615266093-A
Country	US
Kind code	B2
Filing date	Sep 15, 2016
Priority date	Mar 29, 2010
Publication date	Jul 25, 2017
Grant date	Jul 25, 2017

Abstract

A system and method, which may be an offline method, extracts relevant image features about listing items in a network-based publication system for enabling image similarity searching of such listing items. When a seller lists an item, an image of the item is uploaded and may be sent to a picture processing service, which generates several digests. The digests are compressed data structures each representing a particular image feature such as edge, color, texture, or words. These digests are then stored in a search database, where the digests can be used to retrieve listings by image similarity at scale. A similar process can be performed for a query listing for searching the search database for items similar to the query listing.

First claim

What is claimed is:

1. A computer-implemented method comprising:
receiving from a client machine an image query for a desired product that is listed for sale;
computing a first digest of the image query;
determining, based on the first digest, textual information that describes the desired product;
using the determined textual information as a search parameter, performing a first database textual similarity search of second digests of products listed for sale, the second digests being computed before receipt of the query from the client machine, and comprising both image features and textual information of the products, to find first images of products that satisfy the search parameter;
providing to the client machine a plurality of the first images of products that result from the first textual similarity search, and at least one first selectable icon that, when selected, causes a request to find more images of products similar to at least one of the plurality of first images of products;
responsive to selection of one of the at least one first selectable icon, and using textual information related to the at least one of the plurality of first images of products, performing a second textual similarity search of the second digests for second images of products similar to the at least one of the plurality of first images of products; and
providing to the client machine a plurality of the second images of products that result from the second database textual similarity search.

2. The method of claim 1 wherein at least one of the plurality of first images of products or at least one of the plurality of the second images of products comprises video content.

3. The method of claim 1 wherein the determined textual information is one of textual information specified by a seller of the product or textual information generated by an ecommerce system.

4. The method of claim 1 wherein the plurality of the second images of products comprise pixels, and similarity of the second images of products to the at least one of the plurality of first images of products is determined by using textual information associated with the at least one of the plurality of first images of products as a search parameter.

5. The method of claim 4 the textual information comprises at least one of product title or product attributes.

6. One or more computer-readable hardware storage device having embedded therein a set of instructions which, when executed by one or more processors of a computer, causes the computer to execute operations comprising:
receiving from a client machine an image query for a desired product that is listed for sale;
computing a first digest of the image query;
determining, based on the first digest, textual information that desired product;
using the determined textual information as a search parameter, performing a first database textual similarity search of second digests of products listed for sale, the second digests being computed before receipt of the query from the client machine, and comprising both image features and textual information of the products, to find first images of products that satisfy the search parameter;
providing to the client machine a plurality of the first images of products that result from the first textual similarity search, and at least one first selectable icon that, when selected, causes a request to find more images of products similar to at least one of the plurality of first images of products;
responsive to selection of one of the at least one first selectable icon, and using textual information related to the at least one of the plurality of first images of products, performing a second textual similarity search of the second digests for second images of products similar to the at least one of the plurality of first images of products; and
providing to the client machine a plurality of the second images of products that result from the second database textual similarity search.

7. The one or more computer-readable hardware storage device of claim 6 wherein at least one of the plurality of first images of products or at least one of the plurality of the second images of products comprises video content.

8. The one or more computer-readable hardware storage device of claim 6 wherein the textual information related to the query is one of textual information specified by a seller of the product or textual information generated by an ecommerce system.

9. The one or more computer-readable hardware storage device of claim 6 wherein the plurality of the second images of products comprise pixels, and similarity of the second images of products to the at least one of the plurality of first images of products is determined by using textual information associated with the at least one of the plurality of first images of products as a search parameter.

10. The one or more computer-readable hardware storage device of claim 9 wherein the textual information comprises at least one of product title or product attributes.

11. A system comprising:
one or more computer processors and storage configured to:
receive from a client machine an image query for a desired product that is listed for sale;
compute a first digest of the image query;
determine, based on the first digest, textual information that describes the desired product;
by use of determined textual information as a search parameter, perform a first database textual similarity search of second digests of products listed for sale, the second digests being computed before receipt of the query from the client machine, and comprising both image features and textual information of the products, to find first images of products that satisfy the search parameter;
provide to the client machine a plurality of the first images of products that result from the first database textual similarity search, and at least one first selectable icon that, when selected, causes a request to find more images of products similar to at least one of the plurality of first images of products;
responsive to selection of one of the at least one first selectable icon, by use of the textual information related to the at least one of the plurality of first images of products, perform a second database textual similarity search of the second digests for second images of products similar to the at least one of the plurality of first images of products; and
provide to the client machine a plurality of the second images of products that result from the second database textual similarity search.

12. The system of claim 11 wherein the textual information is one of text information specified by a seller of the product or textual information generated by an ecommerce system.

13. The system of claim 11 wherein at least one of the plurality of first images of products or at least one of the plurality of the second images of products comprises video content.

14. The system of claim 11 wherein the plurality of the second images of products comprise pixels, and similarity of the second images of products to the at least one of the plurality of first images of products is determined by use of textual information associated with the at least one of the plurality of first images of products as a search parameter.

Assignees

• Ebay Inc

Inventors

• Chittar Naren
• Ghatare Sanjay Pundlkrao
• Mcdonald Ryan
• Roper John
• Schmitz Michael

Classifications

• G06F16/532 Primary

  Query formulation, e.g. graphical querying · CPC title

• G06F16/5838

  using colour · CPC title

• G06T7/90

  Determination of colour characteristics · CPC title

• G06F16/50

  of still image data · CPC title

• G06V10/56

  relating to colour · CPC title