Anchor image identification for vertical video search

Patent metadata

Field	Value
Publication number	US-9430478-B2
Application number	US-201414147200-A
Country	US
Kind code	B2
Filing date	Jan 3, 2014
Priority date	Jun 23, 2011
Publication date	Aug 30, 2016
Grant date	Aug 30, 2016

Abstract

Anchor images and information associated therewith are accumulated during a Web crawling operation. One or more rules are applied to the accumulated candidate anchor images to filter out candidate anchor images that are not appropriate for use as the anchor image for a particular target video. The remaining candidate anchor image is then selected as the anchor image for the particular video.

First claim

What is claimed is:

1. A computer-implemented method for accumulating candidate anchor images, the computer-implemented method comprising performing computer-implemented operations for:
identifying an image link from a list included in a site, by a computer comprising one or more processors, during a Web crawling operation performed by a video search application;
determining, by the computer, whether the image link points to an image that is representative of a video contained in a Web page having a video page pattern;
associating the image link with the Web page, by the computer, if it is determined that the image link points to the image that is representative of the video contained in the Web page; and
storing the image that is representative of the video contained within the Web page in a storage medium in association with the image link and the Web page, wherein the image stored in the storage medium is a candidate anchor image among a plurality of candidate anchor images in the storage medium for anchor image selection.

2. The computer-implemented method of claim 1 , further comprising disregarding the image link if it is determined, by the computer, that the image link does not point to the image that is representative of the video contained in the Web page.

3. The computer-implemented method of claim 1 , wherein the list included in the site comprises a video list included in a video site.

4. The computer-implemented method of claim 3 , wherein the video list is included in a channel of the video site, the channel comprising videos associated with a particular subject matter.

5. The computer-implemented method of claim 1 , wherein the list is created by the site according to a sorting criterion, the sorting criterion comprising one of most viewed, most recent, least recent, highest rated, and lowest rated.

6. The computer-implemented method of claim 1 , wherein the image link is a hyperlink associated with the image that is representative of the video contained within the Web page.

7. The computer-implemented method of claim 1 , further comprising:
receiving a target page for analysis; and
selecting the stored image from the storage medium as an anchor image for the target page.

8. A computer-implemented method for filtering and selecting an anchor image from a set of candidate anchor images, the computer-implemented method comprising performing computer-implemented operations for:
identifying, by a computer comprising one or more processors, a target image from one or more target images in an image list for consideration as an anchor image for a target video page, the anchor image including an image that is selected to represent the target video page;
generating one or more rules for selecting the anchor image based upon one or more factors for a Web site to which the target video page belongs;
applying, by the computer, to the target image, the one or more rules for selecting the anchor image;
determining automatically, by the computer, that the target image is valid as the anchor image for the target video page according to the one or more rules for selecting the anchor image;
based at least in part on the determining, selecting, by the computer, the target image as the anchor image for the target video page; and
storing the target image as the anchor image for the target video page in a storage medium.

9. The computer-implemented method of claim 8 , wherein identifying the target image for consideration as the anchor image for the target video page comprises performing a Web crawling operation, wherein a result of the Web crawling operation identifies the target image as a candidate anchor image for the target video page.

10. The computer-implemented method of claim 8 , wherein applying the one or more rules for selecting the anchor image comprises applying an image dimension rule, wherein the target image is identified as the anchor image for the target video page based at least in part on the target image meeting a minimum image dimension and the target image not exceeding a maximum image dimension.

11. The computer-implemented method of claim 8 , wherein applying the one or more rules for selecting the anchor image comprises applying an image uniform resource locater (“URL”) pattern rule, wherein the target image is identified as the anchor image for the target video page based at least in part on the target image being associated with a URL that conforms to a URL pattern that is included in a white list.

12. The computer-implemented method of claim 8 , wherein applying the one or more rules for selecting the anchor image comprises applying an image URL pattern rule, wherein the target image is associated with a URL and the URL is checked to see whether the URL conforms to a URL pattern that is included in a white list or conforms to a URL pattern that is included in a black list.

13. The computer-implemented method of claim 8 , wherein generating the one or more rules for selecting the anchor image includes verifying the one or more rules.

14. The computer-implemented method of claim 8 , wherein applying the one or more rules for selecting the anchor image comprises applying a rule to determine whether the target image includes text that is the same as text included in a title of the target video page or a video contained therein, wherein the target image is identified as the anchor image for the target video page based at least in part on the target image including text that is the same as the text included in the title of the target video page or the video contained therein.

15. The computer-implemented method of claim 8 , wherein the target image is determined to be valid as the anchor image based at least in part on at least one of the one or more rules for selecting the anchor image not excluding the target image as the anchor image.

16. A computer storage medium having computer readable instructions stored thereupon that, when executed by a computer, cause the computer to:
identify an image link during a Web crawling operation of a video search application;
determine if the image link points to an image in a Web page having a video page pattern;
associate the image link with the Web page based at least in part on determining that the image link points to the Web page;
apply, to a target image associated with the image link, one or more rules for selecting an anchor image, the anchor image including an image that is selected to represent the Web page;
determine automatically that the target image is valid as the anchor image for the Web page according to the one or more rules for selecting the anchor image based at least in part on at least one of the one or more rules for selecting the anchor image not excluding the target image as the anchor image;
based at least in part on determining automatically that the target image is valid as the anchor image for the Web page, select the target image as the anchor image for the Web page; and
store the target image as the anchor image for the Web page in a storage medium.

17. The computer storage medium of claim 16 , wherein the instructions that, when executed by the computer, cause the computer to apply the one or more rules for selecting the anchor image further comprise instructions that, when executed by the computer, cause the computer to apply an image dimension rule, wherein the target image is identified as the anchor image for the Web page based at least in part on the target image meeting a minimum image dimension and the target image not exceeding a maximum image dimension.

Assignees

• Microsoft Technology Licensing Llc

Inventors

• Kong Xiao
• Wang Wei
• Cai Rui
• Li Haifeng
• Sun Yanfeng

Classifications

• G06F17/30864

  Physics · mapped topic

• G06F17/30047 Primary

  Physics · mapped topic

• G06F17/30867

  Physics · mapped topic

• G06F16/9535

  Search customisation based on user profiles and personalisation · CPC title

• G06F16/434 Primary

  using image data, e.g. images, photos, pictures taken by a user · CPC title