APPARATUS AND METHOD FOR REAL-TIME CAMERA TRACKING

• US 20100158352A1
• Filed: 06/23/2009
• Published: 06/24/2010
• Est. Priority Date: 12/22/2008
• Status: Active Grant

• Current Assignee: Electronics and Telecommunications Research Institute
• Original Assignee: Electronics and Telecommunications Research Institute
• Inventors: YU, Jung-Jae, Kim, Jae-Hean, Jeong, Il-Kwon, Kim, Hye-mi
• Granted Patent: US 8,265,337 B2
• Time in Patent Office: Days
• US Class Current: 382/154
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments
• CPC Class Codes: G06T 7/277 involving stochastic approaches, e.g. using Kalman filters
• CPC Class Codes: H04N 5/2224 related to virtual studio applications

Abstract

A camera tracking apparatus for calculating in real time feature information and camera motion information based on an input image includes a global camera tracking unit for computing a global feature map having feature information on entire feature points; a local camera tracking unit for computing in real time a local feature map having feature information on a part of the entire feature points; a global feature map update unit for receiving the computed feature information from the global and local camera tracking units to update the global feature map; and a local feature selection unit for receiving the updated feature information from the global feature map update unit to select in real time the feature points contained in the local feature map. The local camera tracking unit computes the local feature map for each frame, while the global camera tracking unit computes the global feature map over frames.

Claims

1. A camera tracking apparatus for calculating in real time feature information and camera motion information based on an input image, the apparatus comprising:

a global camera tracking unit for computing a global feature map having feature information on entire feature points;

a local camera tracking unit for computing in real time a local feature map having feature information on a part of the entire feature points;

a global feature map update unit for receiving the computed feature information from the global and local camera tracking units to update the global feature map; and

a local feature selection unit for receiving the updated feature information from the global feature map update unit to select in real time the feature points contained in the local feature map.

2. The apparatus of claim 1, wherein the local camera tracking unit computes the local feature map for each frame, and wherein the global camera tracking unit computes the global feature map over frames.

3. The apparatus of claim 1, wherein the local feature selection unit selects the feature points contained in the local feature map based on a current camera angle.

4. The apparatus of claim 3, wherein the local feature selection unit selects, as the feature points contained in the local feature map, feature points within or adjacent to the current camera angle.

5. The apparatus of claim 3, wherein a specific number of the feature points adjacent to each other forms a feature group;

and wherein the local feature selection unit selects, as the feature points contained in the local feature map, feature points in feature groups within or adjacent to the current camera angle.

6. The apparatus of claim 5, wherein the local feature selection unit selects, as the feature points contained in the local feature map, feature points in each feature group having a centroid within or adjacent to the current camera angle.

7. The apparatus of claim 1, wherein the global feature map has the feature information having been contained in the local feature map in past or being contained in the local feature map currently.

8. The apparatus of claim 7, wherein the feature information and camera motion information contained in the global feature map are represented as a single state vector.

9. The apparatus of claim 8, wherein the feature information includes pixel coordinates in the image and depth information in an axial direction of a camera.

10. The apparatus of claim 8, wherein the state vector is obtained via estimation using a uniform motion model and feature tracking using active feature measurement.

11. A camera tracking method for calculating in real time feature information and camera motion information based on an input image, the method comprising:

computing a global feature map having feature information on entire feature points;

computing in real time a local feature map having feature information on a part of the entire feature points;

updating the global feature map by using the computation result for the global and local feature maps; and

selecting in real time the feature points contained in the local feature map by using the updated feature information.

12. The method of claim 11, wherein the local feature map is computed for each frame, and wherein the global feature map is computed over frames.

13. The method of claim 11, wherein the feature points contained in the local feature map are selected based on a current camera angle.

14. The method of claim 13, wherein feature points within or adjacent to the current camera angle are selected as the feature points contained in the local feature map.

15. The method of claim 13, wherein a specific number of the feature points adjacent to each other forms a feature group;

and wherein feature points in feature groups within or adjacent to the current camera angle are selected as the feature points contained in the local feature map.

16. The method of claim 15, wherein feature points in each feature group having a centroid within or adjacent to the current camera angle are selected as the feature points contained in the local feature map.

17. The method of claim 11, wherein the global feature map has the feature information having been contained in the local feature map in past or being contained in the local feature map currently.

18. The method of claim 17, wherein the feature information and camera motion information contained in the global feature map are represented as a single state vector.

19. The method of claim 18, wherein the feature information includes pixel coordinates in the image and depth information in an axial direction of a camera.

20. The method of claim 18, wherein the state vector is obtained via estimation using a uniform motion model and feature tracking using active feature measurement.