Method and device for creating semantic browsing options

• US 20050246336A1
• Filed: 05/23/2003
• Published: 11/03/2005
• Est. Priority Date: 05/30/2002
• Status: Active Grant

• Current Assignee: InterDigital Madison Patent Holdings (InterDigital, Inc.)
• Original Assignee: Thomson Licensing (Vantiva SA)
• Inventors: Chupeau, Bertrand, Stauder, Jurgen, Nunez, Loic
• Granted Patent: US 8,510,676 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

Abstract

The invention concerns a device for creating semantic browsing options comprising a means to select an initial set of images from an input database comprising a plurality of images, each image being associated with a semantic information out of a semantic information space, a means to determine a semantic information sub-space represented by the semantic information associated with said initial set of images, The device comprises also: a means to select a reduced sub-space out of said sub-space, a means to create semantic links to images of the database associated with the semantic information of said reduced sub-space, a means to display at least one image of the database associated with the semantic information of said reduced sub-space.

Claims

1. Device for creating semantic browsing options comprising

a means to select an initial set of images from an input database comprising a plurality of images, each image being associated with a semantic information out of a semantic information space,

a means to determine a semantic information sub-space represented by the semantic information associated with said initial set of images, wherein the method further comprises

a means to select a reduced sub-space out of said sub-space,

a means to create semantic links to images of the database associated with the semantic information of said reduced sub-space,

a means to display images of the database associated with the semantic information of said reduced sub-space,

a means to select one of said displayed image, triggering the display of another set of images replacing the initial set of images and enabling an iterative browsing of the input database.

2. Device according to claim 1 wherein the means to select an initial set of images can be chosen among

a fixed burned-in selection,

a random selection,

a systematic selection representing the database content,

an interactive selection using a graphical user interface,

a selection that considers a user profile.

3. Device according to claim 1 wherein the input database is chosen among a set of still images, a set of images extracted from at least one video sequence, and a set of images received through a communication link.

4. Device according to claim 1, wherein the means to select a reduced sub-space out of said sub-space are intended to select said reduced sub-space according to a predetermined criterion.

5. Device according to claim 4 wherein the means to select a reduced sub-space out of said sub-space are intended to select said reduced sub-space according to a predetermined criterion chosen according to a user profile.

6. Device according to claim 4, wherein said means to select a reduced sub-space out of said sub-space are intended to modify said predetermined criterion upon interaction of a user.

7. Device according to claim 6 wherein the user interaction consists in selecting at least one of the images of the initial set of images and in that the means to select a reduced sub-space select a reduced sub-space represented by the semantic information associated with said selected images.

8. Device according to claim 6 wherein the user interaction consists in selecting at least one of the images of the initial set of images and in that the means are intended to select a reduced sub-space represented by the semantic information associated with the initial set of images minus the semantic information associated with the selected images.

9. Device according to claim 1, wherein the means to select a reduced sub-space out of said sub-space are intended to select said reduced sub-space by a method selected out of a cluster analysis, a principal component analysis and an histogram analysis of the semantic information of said sub-space.

10. Device according to claim 1, wherein entities chosen among labels of semantic classes, words and word combinations define the semantic information space.

11. Device according to claim 1, wherein the reduced sub-space contains at least one dimension of said semantic information sub-space, said device comprises

means to choose at least one of the dimensions of said sub-space,

means to select for each chosen dimension at least one image of the database where the associated semantic information of this image is strong in the chosen dimension.

12. Device according to claim 11 wherein the means to select a reduced sub-space out of said sub-space are intended to use a graph that contains semantic links between the dimensions of the semantic information space.

13. Device according to claim 11, wherein

the means to display at least one image of the database associated with the semantic information of said reduced sub-space are intended to display the images in the form of image buttons, where the selection of each image button offers a browsing option for images of the database of which associated semantic information is strong in the same dimension as that one associated with the image button.

14. Device according to claim 1, wherein the selected initial set of images and the images of the database associated with the semantic information of said reduced sub-space are displayed simultaneously using at least two windows of a display.

15. Method for creating semantic browsing options comprising the steps of:

selecting an initial set of images from an input database comprising a plurality of images, each image being associated with a semantic information out of a semantic information space,

determining a semantic information sub-space represented by the semantic information associated with said initial set of images,

wherein the method further comprises the steps of

selecting a reduced sub-space out of said sub-space,

creating semantic links to images of the database associated with the semantic information of said reduced sub-space,

displaying at least one image of the database associated with the semantic information of said reduced sub-space.

said method being preferably intended to be implemented in a device according to claim 1.

16. Computer program product comprising program instructions for executing the steps of the method for creating semantic browsing options according to claim 15, when said program is loaded on a computer.