Method for stabilizing a first sequence of digital image frames and image stabilization unit

1. A method for stabilizing a first sequence of digital image frames comprising:

determining a dominant motion vector of a dominant motion layer of said first sequence;

determining a first part of the dominant motion vector representative of an intended motion in said first sequence;

determining a second part of the dominant motion vector representative of an unintended motion in said first sequence;

generating a second sequence from the first sequence based on the second part, the dominant motion vector being determined based on a weighted combination of all vectors that constitute the dominant motion layer;

refining a current estimate of the dominant motion layer by weighting out motion vectors that are beyond a predetermined distance from the dominant motion vector, the dominant motion layer being represented as a matrix including values between a first value and a second value, the first value in the matrix indicating an outlier, and the second value in the matrix indicating an inlier;

setting a third value in the matrix to the first value when the third value is below a first predetermined threshold; and

setting a fourth value in the matrix to the second value when the fourth value is above a second predetermined threshold.

2. The method according to claim 1 wherein said determining the first part includes

applying a low pass filter to the dominant motion vector.

3. The method according to claim 2 , wherein said applying the low pass filter includes

applying a moving average to the dominant motion vector.

4. An image stabilization apparatus for stabilizing a first sequence of digital image frames comprising:

a processor configured:

to determine a dominant motion vector of a dominant motion layer of said first sequence;

to determine a first part of the dominant motion vector representative of an intended motion in said first sequence;

to determine a second part of the dominant motion vector representative of an unintended motion in said first sequence;

to refine a current estimate of the dominant motion layer by weighting out motion vectors that are beyond a predetermined distance from the dominant motion vector, the dominant motion layer being represented as a matrix including values between a first value and a second value, the first value in the matrix indicating an outlier, and the second value in the matrix indicating an inlier;

to set a third value in the matrix to the first value when the third value is below a first predetermined threshold; and

to set a fourth value in the matrix to the second value when the fourth value is above a second predetermined threshold; and

a frame renderer configured to generate a second sequence from the first sequence based on the second part,

wherein the dominant motion vector is determined based on a weighted combination of all vectors that constitute the dominant motion layer.

5. A non-transitory computer-readable storage medium including computer-program instructions, which when executed by a computer, cause the computer to perform a method comprising:

determining a dominant motion vector of a dominant motion layer of a first sequence of digital images;

determining a first part of the dominant motion vector representative of an intended motion in said first sequence;

determining a second part of the dominant motion vector representative of an unintended motion in said first sequence;

generating a second sequence from the first sequence based on the second part, the dominant motion vector being determined based on a weighted combination of all vectors that constitute the dominant motion layer;

refining a current estimate of the dominant motion layer by weighting out motion vectors that are beyond a predetermined distance from the dominant motion vector, the dominant motion layer being represented as a matrix including values between a first value and a second value, the first value in the matrix indicating an outlier, and the second value in the matrix indicating an inlier;

setting a third value in the matrix to the first value when the third value is below a first predetermined threshold; and

setting a fourth value in the matrix to the second value when the fourth value is above a second predetermined threshold.

6. The method according to claim 1 ,

wherein the first value is 0 and the second value is 1,

wherein the first predetermined threshold is 0.25, and

wherein the second predetermined threshold is 0.75.

7. The method according to claim 1 ,

wherein the first predetermined threshold and the second predetermined threshold are values between the first value and the second value.