Searching for more products like a specified product

Patent metadata

Field	Value
Publication number	US-9405773-B2
Application number	US-201113073926-A
Country	US
Kind code	B2
Filing date	Mar 28, 2011
Priority date	Mar 29, 2010
Publication date	Aug 2, 2016
Grant date	Aug 2, 2016

Abstract

Using a processor, receiving, a query including a query image. A database is searched for a set of images similar to the query image, using a two-pass search. The results of the searching are then provided, the results including image members of the set of images similar to the query image. The first pass may be performed using a TF-IDF algorithm and the second pass ranks a predetermined number of results from the first pass by a best match algorithm or other type of algorithm. The type of second pass algorithm may be selectable by a user. If an end signal is not detected, a further two-pass search may be made based on a subsequent query having a subsequent query image. The subsequent query image may be one of the results of the previous two pass search. A subsequent query image may change from an image of a product of a first product category to an image of a product of a second product category.

First claim

What is claimed is:

1. A computer implemented method comprising:
receiving a query comprising a query image that is sent via a user interface at a client machine operating over a wireless communication channel;
responsive to receiving the query image, searching a database and obtaining from the database a set of images that are similar to the query image,
the searching a database and obtaining a set of images that are similar to the query image comprising a two-pass search, a first pass resulting in a plurality of images that are similar to the query image and ranking the plurality of images that are similar to the query image to form a similarity score, and a second pass comprising ranking fewer than all of the resulting plurality of images from the first pass against pre-computed digests of image-based listings comprising textual information, wherein the ranking of fewer than all of the resulting plurality of images is performed using a best match procedure that comprises weighting the resulting plurality of images by color similarity, by edge similarity, and by the similarity score;

providing, over the wireless communication channel for display at the user interface at the client machine, results of the searching, the results comprising image members of the obtained set of images that are similar to the query image;
detecting a request, sent via the user interface at the client machine, for a similarity search for more images similar to one of the image members of the set of images displayed at the user interface at the client machine;
responsive to detecting the request sent via the user interface at the client machine for the similarity search, searching for more images similar to one of the image members of the set of images displayed at the user interface at the client machine, searching the database and obtaining from the database the more images similar to the one of the image members of the set of images; and
responsive to the searching the database and the obtaining from the database more images similar to the one of the image members of the set of images, providing, over the wireless communication channel for display at the user interface at the client machine, at least some of the obtained more images similar to the one of the image members of the set of images.

2. The method of claim 1 , wherein weighting the resulting plurality of images by color similarity comprises performing the second pass by ranking fewer than all of the resulting plurality of images based on a distance between color histograms of the resulting plurality of images that have color attributes.

3. The method of claim 1 , wherein weighting the resulting plurality of images by edge similarity comprises performing the second pass by ranking fewer than all of the resulting plurality of images based on cosine similarity of edge matrices of ones of the fewer than all of the resulting plurality of images that have edge attributes that are computed using edge matrices.

4. The method of claim 1 , wherein the two-pass search comprises performing the second pass by ranking fewer than all of the resulting plurality of images using a pattern comparison of ones of the fewer than all of the resulting plurality of images that have pattern attributes.

5. The method of claim 1 , wherein using the two-pass search comprises performing the second pass by ranking fewer than all of the resulting plurality of images based on text similarity of ones of the fewer than all of the resulting plurality of images that have text attributes.

6. The method of claim 1 , wherein at least one of the image members comprises sale attributes, the method further comprising ranking fewer than all of the image members by at least one of the sale attributes.

7. The method of claim 6 , wherein the sale attributes comprise one of time (sale ending soonest); newly listed; price+shipping cost (lowest first); price+shipping cost (highest first); price (highest first); price (lowest first), seller or item quality.

8. The method of claim 1 , further comprising
determining whether an end signal is received; and
based on the end signal not being received, searching the database and obtaining from the database a set of images similar to a subsequent query image received via the user interface at the client machine.

9. The method of claim 8 wherein the query image comprises an image of a product of a first product category and the subsequent query image comprises an image of a product of a second product category.

10. The method of claim 1 wherein the query comprises structured data and unstructured data.

11. The method of claim 1 wherein the searching the database and obtaining from the database a set of images that are similar to the query image comprises searching in a first pass and a second pass, wherein the second pass is selectable via the user interface at the client machine.

12. The method of claim 11 wherein the first pass is made using a TF-IDF algorithm.

13. The method of claim 1 , the second pass of the two pass search comprising ranking images that result from the first pass of the two-pass search against pre-computed digests of image-based listings comprising textual information, the ranking performed using feedback from users.

14. The method of claim 1 wherein the displayed results comprising image members of the obtained set of images similar to the query image comprise selectable icons that, when selected, comprise the request by the user at the client machine for a similarity search for more images similar to one of the image members of the set of images.

15. A 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 a query comprising a query image that is sent via a user interface at a client machine operating over a wireless communication channel;
responsive to receiving the query image, searching a database and obtaining from the database a set of images that are similar to the query image,
the searching a database and obtaining a set of images that are similar to the query image comprising a two-pass search, a first pass resulting in a plurality of images that are similar to the query image and ranking the plurality of images that are similar to the query image to form a similarity score, and a second pass comprising ranking fewer than all of the resulting plurality of images from the first pass against pre-computed digests of image-based listings comprising textual information, wherein the ranking of fewer than all of the resulting plurality of images is performed using a best match procedure that comprises weighting the resulting plurality of images by color similarity, by edge similarity, and by the similarity score;

providing, over the wireless communication channel, for display at the user interface at the client machine, results of the searching, the results comprising image members of the obtained set of images that are similar to the query image;
detecting a request via the user interface at the client machine for a similarity search for more images similar to one of the image members of the set of images displayed at the user interface at the client machine;
responsive to detecting the request via the user interface at the client machine for the similarity searching for more images similar to one of the image members of the set of images displayed at the user interface at the client machine, searching the database and obtaining from the database more images similar to the one of the image membe

Assignees

• Chittar Naren
• Ghatare Sanjay Pundlkrao
• Henderson Richard D
• Mcdonald Ryan
• Roper John
• Ebay Inc

Inventors

• Chittar Naren
• Ghatare Sanjay Pundlkrao
• Henderson Richard D
• Mcdonald Ryan
• Roper John

Classifications

• G06F17/3025

  Physics · mapped topic

• G06F17/30256

  Physics · mapped topic

• G06F17/30277 Primary

  Physics · mapped topic

• G06F16/532 Primary

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

• G06F16/5838

  using colour · CPC title