Method and mechanism for retrieving images

• US 20070097420A1
• Filed: 12/22/2005
• Published: 05/03/2007
• Est. Priority Date: 10/31/2005
• Status: Active Grant

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Shah, Shesha, Sathish, Sriram
• Granted Patent: US 7,831,111 B2
• Time in Patent Office: Days
• US Class Current: 358/1.15
• 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 16/583 using metadata automatically derived from the content

Abstract

A method and apparatus for, in response to receiving a request, retrieving digital images that are associated with one or more keywords and which are similar to a digital image is provided. The request may be processed using a keywords index and a content index, and optionally, a session index that references data about what digital images users have viewed together in a session. A category associated with the identified digital image may also be used in processing the request, e.g., the category associated with the identified digital image may change the weight placed on various content attributes used in comparing the content of the identified digital image to other digital images. After the set of requested digital images are determined, request data, that identifies the set of requested digital images, may be sent to the requestor.

Claims

1. A method for retrieving digital images, comprising:

generating a content index comprising a plurality of entries, wherein each of the plurality of entries contains image description data that reflects one or more content attributes of one of a plurality of images;

receiving, from a client, request data that requests one or more requested images that are each (a) associated with one or more keywords, and (b) similar to a base image;

determining one or more potential images that are associated with said one or more keywords;

analyzing the base image to determine content data, wherein the content data reflects one or more content attributes of the base image;

determining the one or more requested images by;

for each image in at least a subset of the one or more potential images, comparing the image description data, associated with an entry of the content index for that image, with the content data, thereby determining how similar the content of each image is to the content of the base image, and

selecting a portion of the subset of the one or more potential images as the one or more requested images; and

transmitting, to the client, response data that identifies the one or more requested images.

2. The method of claim 1, wherein the response data includes the one or more requested images.

3. The method of claim 1, further comprising:

determining a particular category, of a plurality of categories, associated with the base image,

wherein the step of determining the one or more requested images further comprises;

determining the subset of the one or more potential images, wherein each image in the subset is associated with the particular category.

4. The method of claim 3, wherein each category, of the plurality of categories, is associated with a set of attribute weight values, and wherein each of the set of attribute weight values describes how much weight a particular content attribute should be assigned when comparing the image description data to the content data.

5. The method of claim 1, wherein determining the one or more potential images comprises:

for each of the one or more keywords, accessing an entry, of a plurality of keyword entries, of a keyword index,

wherein each of the plurality of keyword entries identifies which of the plurality of images is associated with a particular keyword.

6. The method of claim 1, wherein determining the one or more requested images further comprises:

adding a new image to the subset or removing an existing image from the subset based upon session data,

wherein the session data identifies images which one or more users viewed together during a single session of a specified duration.

7. The method of claim 1, wherein determining the one or more requested images is performed, at least in part, on session data, wherein the session data identifies images which one or more users viewed together during a single session of a specified duration.

8. The method of claim 7, wherein the specified duration expires at one or more of:

(a) a time interval, and (b) receipt of another request, from the client, for requested images.

9. The method of claim 7, wherein the session data is organized in a session index, wherein each entry of the session index indicates, for a particular image, other images which the one or more users viewed together with the particular image during the single session of a specified duration, and determining the one or more requested images further comprises:

for each image at in least the subset of the one or more potential images, accessing an entry of the session index, associated with the image, to determine other images which other users viewed in a single session of a specified duration with the image.

10. The method of claim 1, wherein the server stores keyword data that associates keywords with the base image, wherein the request data identifies the one or more keywords by identifying the base image, wherein the request data does not contain the one or more keywords, and wherein the method further comprises:

the server determining the one or more keyword based upon the keyword data.

11. The method of claim 1, wherein the request data identifies a set of criteria, and wherein the set of the one or more potential images is created by removing those images in the one or more potential images which do not match the set of criteria.

12. The method of claim 1, wherein determining the one or more requested images further comprises:

ranking each image in the subset based on how similar the content of each image is to the content of the base image, thereby creating one or more ranked images.

13. A machine-readable medium carrying one or more sequences of instructions for retrieving digital images, wherein execution of the one or more sequences of instructions by one or more processors cause:

generating a content index comprising a plurality of entries, wherein each of the plurality of entries contains image description data that reflects one or more content attributes of one of a plurality of images;

receiving, from a client, request data that requests one or more requested images that are each (a) associated with one or more keywords, and (b) similar to a base image;

determining one or more potential images that are associated with said one or more keywords;

analyzing the base image to determine content data, wherein the content data reflects one or more content attributes of the base image;

determining the one or more requested images by;

for each image in at least a subset of the one or more potential images, comparing the image description data, associated with an entry of the content index for that image, with the content data, thereby determining how similar the content of each image is to the content of the base image, and

selecting a portion of the subset of the one or more potential images as the one or more requested images; and

transmitting, to the client, response data that identifies the one or more requested images.

14. The machine-readable medium of claim 13, wherein the response data includes the one or more requested images.

15. The machine-readable medium of claim 13, wherein execution of the one or more sequences of instructions by the one or more processors further cause:

determining a particular category, of a plurality of categories, associated with the base image,

wherein the step of determining the one or more requested images further comprises;

determining the subset of the one or more potential images, wherein each image in the subset is associated with the particular category.

16. The machine-readable medium of claim 15, wherein each category, of the plurality of categories, is associated with a set of attribute weight values, and wherein each of the set of attribute weight values describes how much weight a particular content attribute should be assigned when comparing the image description data to the content data.

17. The machine-readable medium of claim 13, wherein determining the one or more potential images comprises:

for each of the one or more keywords, accessing an entry, of a plurality of keyword entries, of a keyword index,

wherein each of the plurality of keyword entries identifies which of the plurality of images is associated with a particular keyword.

18. The machine-readable medium of claim 13, wherein determining the one or more requested images further comprises:

adding a new image to the subset or removing an existing image from the subset based upon session data,

wherein the session data identifies images which one or more users viewed together during a single session of a specified duration.

19. The machine-readable medium of claim 13, wherein determining the one or more requested images is performed, at least in part, on session data, wherein the session data identifies images which one or more users viewed together during a single session of a specified duration.

20. The machine-readable medium of claim 19, wherein the specified duration expires at one or more of:

(a) a time interval, and (b) receipt of another request, from the client, for requested images.

21. The machine-readable medium of claim 19, wherein the session data is organized in a session index, wherein each entry of the session index indicates, for a particular image, other images which the one or more users viewed together with the particular image during the single session of a specified duration, and determining the one or more requested images further comprises:

for each image at in least the subset of the one or more potential images, accessing an entry of the session index, associated with the image, to determine other images which other users viewed in a single session of a specified duration with the image.

22. The machine-readable medium of claim 13, wherein the server stores keyword data that associates keywords with the base image, wherein the request data identifies the one or more keywords by identifying the base image, wherein the request data does not contain the one or more keywords, and wherein the method further comprises:

the server determining the one or more keyword based upon the keyword data.

23. The machine-readable medium of claim 13, wherein the request data identifies a set of criteria, and wherein the set of the one or more potential images is created by removing those images in the one or more potential images which do not match the set of criteria.

24. The machine-readable medium of claim 13, wherein determining the one or more requested images further comprises:

ranking each image in the subset based on how similar the content of each image is to the content of the base image, thereby creating one or more ranked images.

25. An apparatus for retrieving digital images, comprising:

a machine-readable medium carrying one or more sequences of instructions; and

one or more processors, wherein execution of the one or more sequences of instructions by the one or more processors causes;

generating a content index comprising a plurality of entries, wherein each of the plurality of entries contains image description data that reflects one or more content attributes of one of a plurality of images;

receiving, from a client, request data that requests one or more requested images that are each (a) associated with one or more keywords, and (b) similar to a base image;

determining one or more potential images that are associated with said one or more keywords;

analyzing the base image to determine content data, wherein the content data reflects one or more content attributes of the base image;

determining the one or more requested images by;

for each image in at least a subset of the one or more potential images, comparing the image description data, associated with an entry of the content index for that image, with the content data, thereby determining how similar the content of each image is to the content of the base image, and

selecting a portion of the subset of the one or more potential images as the one or more requested images; and

transmitting, to the client, response data that identifies the one or more requested images.

26. The apparatus of claim 25, wherein the response data includes the one or more requested images.

27. The apparatus of claim 25, wherein execution of the one or more sequences by the one or more processors further cause:

determining a particular category, of a plurality of categories, associated with the base image,

wherein the step of determining the one or more requested images further comprises;

determining the subset of the one or more potential images, wherein each image in the subset is associated with the particular category.

28. The apparatus of claim 27, wherein each category, of the plurality of categories, is associated with a set of attribute weight values, and wherein each of the set of attribute weight values describes how much weight a particular content attribute should be assigned when comparing the image description data to the content data.

29. The apparatus of claim 25, wherein determining the one or more potential images comprises:

for each of the one or more keywords, accessing an entry, of a plurality of keyword entries, of a keyword index,

wherein each of the plurality of keyword entries identifies which of the plurality of images is associated with a particular keyword.

30. The apparatus of claim 25, wherein determining the one or more requested images further comprises:

adding a new image to the subset or removing an existing image from the subset based upon session data,

wherein the session data identifies images which one or more users viewed together during a single session of a specified duration.

31. The apparatus of claim 25, wherein determining the one or more requested images is performed, at least in part, on session data, wherein the session data identifies images which one or more users viewed together during a single session of a specified duration.

32. The apparatus of claim 31, wherein the specified duration expires at one or more of:

(a) a time interval, and (b) receipt of another request, from the client, for requested images.

33. The apparatus of claim 31, wherein the session data is organized in a session index, wherein each entry of the session index indicates, for a particular image, other images which the one or more users viewed together with the particular image during the single session of a specified duration, and determining the one or more requested images further comprises:

for each image at in least the subset of the one or more potential images, accessing an entry of the session index, associated with the image, to determine other images which other users viewed in a single session of a specified duration with the image.

34. The apparatus of claim 25, wherein the server stores keyword data that associates keywords with the base image, wherein the request data identifies the one or more keywords by identifying the base image, wherein the request data does not contain the one or more keywords, and wherein the method further comprises:

the server determining the one or more keyword based upon the keyword data.

35. The apparatus of claim 25, wherein the request data identifies a set of criteria, and wherein the set of the one or more potential images is created by removing those images in the one or more potential images which do not match the set of criteria.

36. The apparatus of claim 25, wherein determining the one or more requested images further comprises:

ranking each image in the subset based on how similar the content of each image is to the content of the base image, thereby creating one or more ranked images.