Image search using face detection

• US 8,285,713 B2
• Filed: 09/08/2008
• Issued: 10/09/2012
• Est. Priority Date: 09/30/2007
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Luo, Jing, Zhang, Xiao Xun, Su, Zhong
• Primary Examiner(s): Ehichioya, Fred I
• Assistant Examiner(s): Fan, Shiow-Jy
• Application Number:
  US12/206,147
• Publication Number:
  US 20090234842A1
• Time in Patent Office: 1,492 Days
• Field of Search: 707/999.005, 707/E17.014, 707/E17.03, 707/722, 707/758, 707/723
• US Class Current: 707/723
• 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 method for searching images using face detection, the method comprising:

executing with a processor of an information processing system the following;

receiving a query submitted by a user, wherein the query comprises at least one query word;

searching, in response to receiving the query, a set of image resources;

obtaining, in response to the searching, an initial set of images from the set of image resources that is associated with the query word;

identifying a set of images in the initial set of images that comprises at least one face, the set of images comprising a first order;

analyzing a set of text comprising the query word and associated with each image in the set of images that has been identified;

determining, in response to the analyzing, if the query word within the set of text indicates a person'"'"'s name;

in response to the query word indicating a person'"'"'s name, analyzing the query to obtain a number, M, of people'"'"'s names in the query;

counting a number, N, of faces in each image in the set of images that have been identified;

dividing the initial set of images into a set of three groups of images comprising a first group of M=N, a second group of M<

N, and a third group of M>

N;

re-ordering the initial set of images in a second order based on the set of three groups with the group M=N being a first ordered group in the second order; and

returning one or more images in the initial set of images that have been re-ordered in the second order in response to the query word in the set of text associated with the one or more images indicating a person'"'"'s name.

2. The method according to claim 1, wherein determining if the query word within the set of text indicates a person'"'"'s name further comprises:

using a Named Entity Recognition technology to determine if the query word within the set of text indicates a person'"'"'s name.

3. The method according to claim 1, further comprising:

using a Named Entity Recognition technology to determine if the query comprises at least one person'"'"'s name.

4. The method according to claim 1, wherein the set of image resources is at least one of:

an intranet,

a local area network,

a wide area network, and

an Internet.

5. The method according to claim 1, wherein returning one or more images further comprises:

displaying each image in the initial set of images that have been re-ordered in the second order according to an order of the first group, the second group, and the third group within the second order.

6. The method according to claim 5, wherein returning the one or more images further comprises displaying one or more images failing to comprise a face.

7. An image search system using face detection, comprising:

a memory;

a processor; and

a face searching system communicatively coupled to the memory and the processor, the face searching system configured to perform a method comprising;

receiving a query submitted by a user, wherein the query comprises at least one query word;

searching, in response to receiving the query, a set of image resources;

obtaining, in response to the searching, an initial set of images from the set of image resources that is associated with the query word;

identifying a set of images in the initial set of images that comprises at least one face, the set of images comprising a first order;

analyzing a set of text comprising the query word and associated with each image in the set of images that has been identified;

determining, in response to the analyzing, if the query word within the set of text indicates a person'"'"'s name;

in response to the query word indicating a person'"'"'s name, analyzing the query to obtain a number, M, of people'"'"'s names in the query;

counting a number, N, of faces in each image in the set of images that have been identified;

dividing the initial set of images into a set of three groups of images comprising a first group of M=N, a second group of M<

N, and a third group of M>

N;

re-ordering the initial set of images in a second order based on the set of three groups with the group M=N being a first ordered group in the second order; and

returning one or more images in the initial set of images that have been re-ordered in the second order in response to the query word in the set of text associated with the one or more images indicating a person'"'"'s name.

8. The system according to claim 7, wherein the searching further comprises:

using a Named Entity Recognition technology to determine if the query word within the set of text indicates a person'"'"'s name.

9. The system according to claim 7, further comprising:

using a Named Entity Recognition technology to determine if the query comprises at least one person'"'"'s name.

10. The system according to claim 7, wherein the set of image resources is at least one of:

an intranet,

a local area network,

a wide area network, and

an Internet.

11. The system according to claim 7, wherein returning one or more images further comprises:

displaying each image in the initial set of images that have been re-ordered in the second order according to an order of the first group, the second group, and the third group within the second order.

12. The system according to claim 11, wherein returning the one or more images further comprises:

displaying one or more images failing to comprise a face.