Real time vergence control for 3D video capture and display

• US 8,754,929 B1
• Filed: 05/23/2011
• Issued: 06/17/2014
• Est. Priority Date: 05/23/2011
• Status: Active Grant

• Current Assignee: John Prince
• Original Assignee: John Prince
• Inventors: Prince, John
• Primary Examiner(s): Lee, Y
• Assistant Examiner(s): SUH, JOSEPH JINWOO
• Application Number:
  US13/068,885
• Time in Patent Office: 1,121 Days
• Field of Search: 348/47, 348/148, 382/284, 455/556.1
• US Class Current: 348/47
• CPC Class Codes: H04N 13/128 Adjusting depth or disparity
• CPC Class Codes: H04N 13/139 Format conversion, e.g. of frame-rate or size

Abstract

This invention relates to the recording and display three-dimensional (3D) images, notably in real-time video capture. This invention focuses on the ability to alter vergence instantaneously through software in order to amend disparities caused by differences between focal length and vergence distance. No mechanical means are necessary for amending both horizontal and vertical disparity. The ability to vary vergence allows selected images to be made to appear behind, level with, or in front of a display screen. While the invention is useful for consumer displays in television and films, the invention may also be used, for example, for visualizing weather, for mapping, for seismic surveys or for aircraft flight control. It is applicable, in real time, throughout the electro-magnetic spectrum particularly in the visible, infra-red and microwave portions.

Claims

1. A method operative on a computer for altering vergence points for multiple detectors in real time, the method comprising:

a pre-alignment step of converging the detectors on a chosen intermediate point by mechanical means, said pre-alignment step including a focal length adjustment of all said detectors;

a secondary alignment step whereby image features from said detectors are identified and used in a transformation to bring all images into substantially equal orientation and size;

a fine alignment step of aligning said images through at least one of a sum of differences squared algorithm or a Fourier transform algorithm;

a vergence step applied in correlated steps to move apparent vergence points evenly to a location nearer to or further away from said detectors;

each said detector image contains image pixels x_i and y_i, thereby forming respective image matrices

2. The method as in claim 1 wherein said vergence step is performed after acquisition of an image from said detector, during said secondary alignment step, or during said fine alignment step.

3. The method as in claim 1 wherein said vergence step is performed in real-time.

4. The method as in claim 1 wherein the number of detectors is two or more.

5. The method as in claim 1 wherein an inverse matrix operation is performed on a previous transformations in real-time through parametric inversion, said inverse matrix operation restoring an original value from the operation of said previous transformation.

6. The method as in claim 5 wherein data streams are encoded through transformation at different time intervals and decoded synchronously using a parametric inversion.

7. The method as in claim 1 wherein said transformations are performed on an off-axis image feature.

8. The method as in claim 1 wherein said transformations are at least one of:

similarity, affine, Euclidean, or non-Euclidean.

9. The method as in claim 8 wherein features are made more prominent or less prominent by said transformation.

10. The method as in claim 8 wherein image disparities are reduced or increased by said transformations.

11. The method as in claim 1 wherein said mechanical pre-alignment step is performed electronically during said fine alignment step.

12. An image processor altering vergence points from multiple detectors, the image processor operative in real time, the image processor:

converging the detectors on a chosen intermediate point by mechanical means and performing a focal length adjustment of all said detectors;

the image processor identifying image features from said detectors, said image features transformed to bring all images into substantially equal orientation and size;

the image processor aligning said images through at least one of a sum of differences squared algorithm or a Fourier transform algorithm;

the image processor moving apparent vergence points evenly to a location nearer to or further away from said detectors;

each said detector image contains image pixels x_i and y_i, thereby forming respective image matrices

13. The image processor of claim 12 where said image processor moving apparent vergence points is performed either after acquisition of images from said detectors or when said image processor is aligning said images.

14. The image processor of claim 12 where moving apparent vergence points is performed by said image processor in real-time.

15. The image processor of claim 12 where the number of said multiple detectors is two.

16. The image processor of claim 12 where said image processor also performs an inverse matrix operation on one or more previous transformations in real-time through parametric inversion, said inverse matrix operation restoring an original value from the operation of said previous transformation.

17. The image processor of claim 16 where said detectors generate data streams which are encoded through transformation at different time intervals and decoded synchronously using said parametric inversion.

18. The image processor of claim 12 where said transformations are performed on an image feature which is off-axis.

19. The image processor of claim 12 where said transformation is at least one of a similarity transformation, an affine transformation, a Euclidean transformation, or a non-Euclidiean transformation.

20. The image processor of claim 12 where said transformation reduces or increases an image disparity.