IMAGE SEARCH USING FACE DETECTION

• US 20090234842A1
• Filed: 09/08/2008
• Published: 09/17/2009
• Est. Priority Date: 09/30/2007
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: LUO, JING, Zhang, Xiao Xun, Su, Zhong
• Granted Patent: US 8,285,713 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• 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

An image search method and system using face detection, begins with receiving a query submitted by a user. Next a query word is searched in the query from an image resource using an image search engine to obtain an initial image collection. Any faces are detected in each image in the initial image collection which has been searched. A search for the query word in a text surrounding each image having the face in the initial image collection is performed. A determination is made whether the query word indicates at least one person'"'"'s name in the surrounding text matching the query word. An image an image in the initial image collection is returned to a user in which the face is included and the query word in the surrounding text indicates the person'"'"'s name.

Claims

1. A computer-implemented method for performing an image search, the method comprising:

receiving a query submitted by a user;

searching for a query word in the query from an image resource using an image search engine to obtain an initial image collection;

detecting a face in each image in the initial image collection which has been searched;

searching for the query word in a text surrounding each image having the face in the initial image collection;

determining whether the query word indicates at least one person'"'"'s name in the text surrounding each image matching the query word; and

returning an image in the initial image collection to a user in which the face is included and the query word in the text surrounding each image indicates the person'"'"'s name.

2. The method according to claim 1, wherein the determining whether the query word indicates the person'"'"'s name in the text surrounding each image matching the query word includes using Named Entity Recognition technology is used to determine whether the query word in the surrounding text indicates the person'"'"'s name.

3. The method according to claim 2, wherein the receiving the query submitted by the user includes using Named Entity Recognition technology to determine whether the query includes the person'"'"'s name.

4. The method according to claim 3, wherein

the image resource is at least one of an intranet, a local area network, a wide area a network, and the Internet.

5. The method according to claim 3, further comprising:

an analyzing step of analyzing the query to obtain a number M of people names in the query.

6. The method according to claim 5, further comprising:

counting a number N of faces which has been detected in each image in which the faces are detected.

7. The method according to claim 6, further comprising:

dividing the initial image collection into three groups according to M=N, M<

N and M>

N.

8. The method according to claim 7, further comprising:

reordering an order of at least one image in the initial image collection in each group and arranging each image with N=0 at an end of the group.

9. The method according to claim 8, wherein the returning the image in the initial image collection to the user includes displaying each image in the image collection according to the order after being reordered.

10. The method according to claim 9, wherein the returning the image in the initial image collection to the user includes displaying only each image having faces.

11. An image search system, the system comprising:

a receiving means for receiving a query submitted by a user;

a first searching means for searching for a query word in the query from an image resource using an image search engine to obtain an initial image collection;

a detecting means for detecting a face in each image in the initial image collection which has been searched;

a second searching means for searching for the query word in a text surrounding each image having the face in the initial image collection;

a determining means for determining whether the query word indicates a at least one Person'"'"'s name in the text surrounding each image; and

a returning means for returning an image in the initial image collection to a user in which a face is included and the query word in the text surrounding each image indicates a person'"'"'s name.

12. The system according to claim 11, wherein the second searching means uses Named Entity Recognition technology to determine whether the query word in the text surrounding each image indicates the person'"'"'s name.

13. The system according to claim 12, wherein the receiving means uses Named Entity Recognition technology to judge whether the query includes the person'"'"'s name.

14. The system according to claim 12, wherein the image resource is at least one of an intranet, a local area network, a wide area network, and an Internet.

15. The system according to claim 13, further comprising:

an analyzing means for analyzing the query to obtain a number M of person'"'"'s names in the query.

16. The system according to claim 15, further comprising:

a face counting means for counting a number N of faces which has been detected in each image in which the faces are detected at the detecting means.

17. The system according to claim 16, further comprising:

a dividing means for dividing the initial image collection into three groups according to M=N, M<

N and M>

N.

18. The system according to claim 17, further comprising:

a reordering means for remaining the order of at least one image in the initial image collection in each group and arranging each image with N=0 at an end of the group.

19. The system according to claim 18, wherein the returning means displays each image in the image collection according to an order after being reordered.

20. The system according to claim 19, wherein the returning means displays only each image having faces.