Calculating transformation parameters for image processing
View Patent ↗The invention relates to a method for obtaining transformation parameters. A camera motion can be modeled by providing a vector field of motion vectors describing estimated motion vectors, projecting the vector field on at least one axis, and deriving the transformation vector parameters from the projection of the vector fields. As the camera motion can be modeled by way of translation, scale and rotation, the projections of the vector field on the axis can be used.
1. A method of obtaining transformation parameters from a vector field with an image processing device, the method comprising
receiving a video image from a video source, the video image having consecutive video frames;
obtaining, with a processor, the vector field from the video image;
projecting, with the processor, the vector field on at least one axis;
deriving, with the processor, the transformation parameters from the projection of the vector field;
compressing, with the processor, the video image using the transformation parameters; and
storing the compressed video image on a non-transitory computer-readable medium.
2. The method of claim 1 , wherein the vector field is described by estimated motion vectors describing motion in an image.
3. The method of claim 1 , wherein the vector field comprises vectors with at least one first component and at least one second component, and wherein the first component is parallel to a first axis and wherein the second component is parallel to a second axis.
4. The method of claim 3 , wherein projecting at least one component on at least one axis isolates at least a scale parameter.
5. The method of claim 4 , wherein the scale parameter is calculated from at least one multiplier of the projection(s) on at least one axis.
6. The method of claim 3 , wherein projecting at least one component on at least one axis isolates at least a translation parameter.
7. The method of claim 6 , wherein the translation parameter is calculated from a mean value of the projection on one axis.
8. The method of claim 3 , wherein projecting at least one component on at least one axis isolates at least a rotation parameter.
9. The method of claim 7 , wherein the rotation parameter is calculated from at least one multiplier of the projection(s) on at least one axis.
10. The method of claim 1 , wherein the transformation parameters are obtained from a curve fit algorithm applied to the projections.
11. The method of claim 1 , wherein the transformation parameters are obtained from a primitive functions of the projections.
12. The method of claim 11 , wherein the primitive function is an Euler function.
13. The method of claim 11 , wherein at least one parameter is calculated from an extreme of the primitive function.
14. The method of claim 1 , wherein the translation parameters represent a global model of camera motion.
15. A non-transitory computer-readable medium encoded with data instructions operable to cause a processor, when operated, to obtain transformation parameters, causes a computing device to:
receive a video image from a video source, the video image having consecutive video frames;
obtain a vector field from the video image;
project the vector field on at least one axis;
derive the transformation parameters from the projection of the vector field;
compressing, with the processor, the video image using the transformation parameters; and
storing the compressed video image on the non-transitory computer-readable medium.
16. An apparatus configured to obtain transformation parameters, comprising:
a video input configured to receive a video image from a video source, the video image having consecutive video frames;
a motion estimator within a computer processor configured to obtain a vector field,
a projecting circuit within the computer processor configured to project the vector field on at least one axis;
a calculation circuit within the computer processor, configured to derive the transformation parameters from the projection of the vector field, and compressing the video image using the transformation parameters; and
a non-transitory computer-readable medium configured to store the compressed video image.
17. The method of claim 1 wherein the video source is a camera.
18. The method of claim 15 wherein the video source is a camera.
19. The method of claim 16 wherein the video source is a camera.