Shape based picture search

• US 8,775,401 B2
• Filed: 02/01/2013
• Issued: 07/08/2014
• Est. Priority Date: 03/26/2009
• Status: Active Grant

• Current Assignee: Alibaba Group Holding Ltd.
• Original Assignee: Alibaba Group Holding Ltd.
• Inventors: Zhou, Chunyi, Wang, Weiwei, Zhou, Xinfeng, Dong, Yu, Weng, Xiaoying, Huang, Jialong
• Primary Examiner(s): Wong, Leslie
• Application Number:
  US13/757,479
• Publication Number:
  US 20140046944A1
• Time in Patent Office: 522 Days
• Field of Search: 707/758, 707/915, 707/999.003, 707/999.006, 707/999.107, 707/E17.089
• US Class Current: 707/707
• 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 16/5854 using shape and object relationship
• CPC Class Codes: G06F 16/5866 using information manually generated, e.g. tags, keywords, comments, manually generated location and time information

Abstract

The present application relates to a method for implementing picture search and a website server thereof. A method for implementing picture search includes: classifying, according to keywords in advance in a picture database, corresponding pictures by shape of objects in the pictures, and determining a sample picture for each shape type; wherein, after a server receives a picture search request sent from a client, the method includes: searching, by the server, in the picture database for the sample picture of several shape types classified in advance corresponding to the keywords in said search request, and returning, to the client, the searched sample picture of the several shape types; receiving, by the server, the sample picture of a certain shape type determined by the client, and searching, in the picture database for the pictures which correspond to said keywords and satisfy a predetermined request with the characteristic value of said determined sample pictures; returning, by the server, said found pictures to the client. The present application enables the user to search pictures of similar shapes according to the shape types, thereby satisfying the user'"'"'s search demands.

Claims

1. A system for picture search, comprising:

one or more processors configured to;

receive a picture search request from a user, wherein the picture search request includes a keyword;

identify a first set of found pictures in a picture database, the first set of found pictures corresponding to the keyword in the picture search request;

present a first plurality of sample pictures that corresponds to a plurality of picture shape types to the user, the first plurality of sample pictures being selected from the first set of found pictures, wherein the first plurality of sample pictures is included in a second plurality of sample pictures predetermined based at least in part on;

classifying a plurality of pictures in the picture database according to a plurality of keywords associated with the respective ones of the plurality of pictures and according to the plurality of picture shape types, wherein classifying the plurality of pictures includes, for a first keyword of the plurality of keywords, extracting a set of pictures associated with the first keyword and identifying at least some of the plurality of picture shapes within the extracted set of pictures;

determining the second plurality of sample pictures among the classified plurality of pictures that corresponds to respective ones of the plurality of picture shape types;

receive a determined sample picture that corresponds to a user selection of one of the first plurality of sample pictures; and

identify a second set of found pictures, the second set of found pictures being selected from the first set of found pictures and having a set of corresponding characteristic values that satisfy a predetermined criterion based at least in part on the determined sample picture; and

one or more memories coupled to the one or more processors and configured to provide the one or more processors with instructions.

2. The system of claim 1, wherein the first plurality of sample pictures is presented to the user in a sorted order.

3. The system of claim 1, wherein classifying the plurality of pictures according to the plurality of picture shape types includes comparing a plurality of characteristic values of the plurality of pictures.

4. The system of claim 3, wherein comparing the plurality of characteristic values of the plurality of pictures includes grouping pictures whose differences of characteristic values are within a predetermined distance as pictures of an identical or similar shape types.

5. The system of claim 1, wherein identifying the second set of found pictures includes identifying, among the first set of found pictures, pictures whose distances from a characteristic value vector of the determined sample picture is below a predetermined threshold.

6. The system of claim 1, wherein the second set of found pictures has an identical or a similar shape type to the determined sample picture.

7. The system of claim 1, wherein at least one of the second plurality of sample pictures is marked with an indicator associated with a sample picture.

8. A method for picture search, comprising:

receiving a picture search request from a user, wherein the picture search request includes a keyword;

identifying, using one or more processors, a first set of found pictures in a picture database, the first set of found pictures corresponding to the keyword in the picture search request;

presenting a first plurality of sample pictures that corresponds to a plurality of picture shape types to the user, the first plurality of sample pictures being selected from the first set of found pictures, wherein the first plurality of sample pictures is included in a second plurality of sample pictures predetermined based at least in part on;

classifying a plurality of pictures in the picture database according to a plurality of keywords associated with the respective ones of the plurality of pictures and according to the plurality of picture shape types, wherein classifying the plurality of pictures includes, for a first keyword of the plurality of keywords, extracting a set of pictures associated with the first keyword and identifying at least some of the plurality of picture shapes within the extracted set of pictures;

determining the second plurality of sample pictures among the classified plurality of pictures that corresponds to respective ones of the plurality of picture shape types;

receiving a determined sample picture that corresponds to a user selection of one of the first plurality of sample pictures; and

identifying a second set of found pictures, the second set of found pictures being selected from the first set of found pictures and having a set of corresponding characteristic values that satisfy a predetermined criterion based at least in part on the determined sample picture.

9. The method of claim 8, wherein the first plurality of sample pictures is presented to the user in a sorted order.

10. The method of claim 8, wherein classifying the plurality of pictures according to the plurality of picture shape types includes comparing a plurality of characteristic values of the plurality of pictures.

11. The method of claim 10, wherein comparing the plurality of characteristic values of the plurality of pictures includes grouping pictures whose differences of characteristic values are within a predetermined distance as pictures of an identical or similar shape types.

12. The method of claim 8, wherein identifying the second set of found pictures includes identifying, among the first set of found pictures, pictures whose distances from a characteristic value vector of the determined sample picture are below a predetermined threshold.

13. The method of claim 8, wherein the second set of found pictures has an identical or a similar shape type to the determined sample picture.

14. The method of claim 8, wherein at least one of the second plurality of sample pictures is marked with an indicator associated with a sample picture.

15. A computer program product for picture search, the computer program product being embodied in a non-transitory computer storage medium and comprising computer instructions for:

receiving a picture search request from a user, wherein the picture search request includes a keyword;

identifying a first set of found pictures in a picture database, the first set of found pictures corresponding to the keyword in the picture search request;

presenting a first plurality of sample pictures that corresponds to a plurality of picture shape types to the user, the first plurality of sample pictures being selected from the first set of found pictures, wherein the first plurality of sample pictures is included in a second plurality of sample pictures predetermined based at least in part on;

classifying a plurality of pictures in the picture database according to a plurality of keywords associated with the respective ones of the plurality of pictures and according to the plurality of picture shape types, wherein classifying the plurality of pictures includes, for a first keyword of the plurality of keywords, extracting a set of pictures associated with the first keyword and identifying at least some of the plurality of picture shapes within the extracted set of pictures;

determining the second plurality of sample pictures among the classified plurality of pictures that corresponds to respective ones of the plurality of picture shape types;

receiving a determined sample picture that corresponds to a user selection of one of the first plurality of sample pictures; and

identifying a second set of found pictures, the second set of found pictures being selected from the first set of found pictures and having a set of corresponding characteristic values that satisfy a predetermined criterion based at least in part on the determined sample picture.

16. The computer program product of claim 15, wherein the first plurality of sample pictures is presented to the user in a sorted order.

17. The computer program product of claim 15, wherein classifying the plurality of pictures according to the plurality of picture shape types includes comparing a plurality of characteristic values of the plurality of pictures.

18. The computer program product of claim 17, wherein comparing the plurality of characteristic values of the plurality of pictures includes grouping pictures whose differences of characteristic values are within a predetermined distance as pictures of an identical or similar shape types.

19. The computer program product of claim 15, wherein identifying the second set of found pictures includes identifying, among the first set of found pictures, pictures whose distances from a characteristic value vector of the determined sample picture are below a predetermined threshold.

20. The computer program product of claim 15, wherein the second set of found pictures has an identical or a similar shape type to the determined sample picture.