APPARATUS AND METHOD FOR ANALYZING VIDEO

• US 20120155711A1
• Filed: 12/16/2011
• Published: 06/21/2012
• Est. Priority Date: 12/16/2010
• Status: Abandoned Application

• Current Assignee: Electronics and Telecommunications Research Institute
• Original Assignee: Electronics and Telecommunications Research Institute
• Inventors: RYOO, Sahngwon, YU, Won Pil, LEE, Jae Yeong, CHOI, Sung Lok
• Application Number:
  US13/328,318
• Publication Number:
  US 20120155711A1
• Time in Patent Office: Days
• US Class Current: 382/104
• CPC Class Codes: G06V 20/58 Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
• CPC Class Codes: G07C 5/0866 the electronic data carrier being a digital video recorder in combination with video camera

Abstract

Disclosed are an apparatus and a method for analyzing a video. The video analyzing apparatus includes a route analyzing unit configured to analyze a subject of a first-person view video and an object represented in the first-person view video based on the first-person view video and generate route information of the subject and the object; and an event analyzing unit configured to classify the first-person view video as a semantic event using the route information.

Claims

1. An apparatus of analyzing a video, comprising:

a route analyzing unit configured to analyze a subject of a first-person view video and an object represented in the first-person view video based on the first-person view video and generate route information of the subject and the object; and

an event analyzing unit configured to classify the first-person view video as a semantic event using the route information.

2. The apparatus of claim 1, wherein the semantic event includes sub events in which the events of each of the subject and the object are classified based on a predetermined criterion.

3. The apparatus of claim 2, wherein the semantic event further includes an event that combines the sub events in order to define the interaction between the subject and the object.

4. The apparatus of claim 1, wherein the event analyzing unit includes:

an event storing unit configured to store a first-person view video and route information corresponding to the classified semantic event.

5. The apparatus of claim 1, wherein the route analyzing unit includes:

a route generating unit configured to generate a route of the subject;

an object detecting unit configured to detect the object; and

a route tracking unit configured to map and track the routes of the subject and the object.

6. The apparatus of claim 5, wherein the route generating unit generates information regarding the position of the subject and a ground plane.

7. The apparatus of claim 6, wherein the information regarding the position of the subject is generated by using the relative position of a pattern that is represented in adjacent images of the first-person view video.

8. The apparatus of claim 5, wherein the object detecting unit detects the position of the object from an image of the first-person view video.

9. The apparatus of claim 5, wherein the route tracking unit maps the route of the subject onto a coordinate system, and maps the object on the basis of the coordinate system.

10. The apparatus of claim 5, wherein the route tracking unit compares the information detected from the object detecting unit with a predetermined hypothesis to check the validity of the detected information.

11. A method of analyzing a video, comprising:

a route analyzing step to analyze a subject of a first-person view video and an object represented in the first-person view video based on the first-person view video and generate route information of the subject and the object; and

an event analyzing step to classify the first-person view video as a semantic event using the route information.

12. The method of claim 11, wherein the semantic event includes sub events in which the events of each of the subject and the object are classified based on a predetermined criterion.

13. The method of claim 12, wherein the semantic event further includes an event that combines the sub events in order to define the interaction between the subject and the object.

14. The method of claim 11, wherein the event analyzing step includes:

an event storing step to store a first-person view video and route information corresponding to the classified semantic event.

15. The method of claim 11, wherein the route analyzing step includes:

a route generating step to generate a route of the subject;

an object detecting step to detect the object; and

a route tracking step to map and track the routes of the subject and the object.

16. The method of claim 15, wherein the route generating step generates information regarding the position of the subject and a ground plane.

17. The method of claim 16, wherein the information regarding the position of the subject is generated by using the relative position of a pattern that is represented in adjacent images of the first-person view video.

18. The method of claim 15, wherein the object detecting step detects the position of the object from an image of the first-person view video.

19. The method of claim 15, wherein the route tracking step maps the route of the subject onto a coordinate system, and maps the object on the basis of the coordinate system.

20. The method of claim 15, wherein the route tracking step compares the detected information with a predetermined hypothesis to check the validity of the detected information.