Error resilient rate distortion optimization for image and video encoding

1. A method for distortion optimization of image or video data, comprising:

providing a coded image or video data;

evaluating distortion of the coded image or video data as a function of transmission distortion of the image or video data;

and

performing distortion optimization of the image or video data while encoding the image or video data, wherein

the distortion relates to a pixel distortion after motion compensation, and

evaluating the distortion includes calculating an approximated second moment of said distortion based on the following equation:

E

⁡

[

X

2

]

≈

∑

i

=

0

N

-

1

⁢

w

i

⁢

E

⁡

[

(

x

i

)

2

]

wherein:

X=D d (u,j) denotes a distortion related to a decoder side and being dependent on a pixel position u and a picture index j;

w denotes a weight parameter wherein

∑

i

=

0

N

-

1

⁢

w

i

=

1

,

and

N denotes an integer number of pixel locations.

2. The method as recited in claim 1 , wherein the transmission distortion of the image or video data is a difference between an input image or video data before encoding and a corresponding output image or video data after decoding.

3. The method as recited in claim 2 , wherein the input image or video data before encoding comprises a known value.

4. The method as recited in claim 2 , wherein the output image or video data after decoding comprises an unknown random variable.

5. The method as recited in claim 2 , wherein the transmission distortion comprises a function of channel losses.

6. The method as recited in claim 5 , wherein the distortion evaluation comprises evaluating the distortion in terms of a linear term and a quadratic term, each of the linear term and the quadratic term expressing the transmission distortion of the image or video data.

7. The method as recited in claim 6 , wherein the evaluating the distortion comprises computing a distortion expected at the decoder.

8. The method as recited in claim 7 , wherein the distortion optimization is performed through motion vectors during an error concealment step.

9. An encoder adapted to receive an image or video data, the encoder comprising:

an input for receiving the image or video data;

a reference picture buffer unit, the reference picture buffer unit containing stored image or video data;

a rate distortion optimization unit, the rate distortion optimization unit adapted to receive the image or video data from the input, wherein the rate distortion optimization unit is further adapted to evaluate a distortion and to calculate an approximated second moment of said distortion based on the following equation, the distortion related to a pixel distortion after motion compensation:

E

⁡

[

X

2

]

≈

∑

i

=

0

N

-

1

⁢

w

i

⁢

E

⁡

[

(

X

i

)

2

]

wherein:

X=D d (u,j) denotes a distortion related to a decoder side and being dependent on a pixel position u and a picture index j;

w denotes a weight parameter wherein

∑

i

=

0

N

-

1

⁢

w

i

=

1

,

and

N denotes an integer number of pixel locations;

a motion estimation unit, the motion estimation unit adapted to receive the image or video data from the input and the rate distortion optimization unit, and adapted to receive stored image or video data from the reference picture buffer unit;

a motion compensation unit, the motion compensation unit adapted to receive the image or video data from the motion estimation unit and the stored image or video data from the reference picture buffer unit;

a transforming unit; and

a quantizing unit, the quantizing unit and the transforming unit adapted to receive a combined image or video data from the input and the motion compensation unit.

10. The encoder as recited in claim 9 , wherein the rate distortion optimization unit is adapted to receive feedback from the motion compensation unit.

11. The encoder as recited in claim 10 , wherein the rate distortion optimization unit further comprises:

a first moment of transmission distortion computation unit;

a first moment of transmission distortion buffer unit, the first moment of transmission buffer unit adapted to receive a computed first moment of transmission distortion value from the first moment of transmission distortion computation unit;

a second moment of transmission distortion computation unit;

a second moment of transmission distortion buffer unit, the second moment of transmission distortion buffer unit adapted to receive a computed second moment of transmission distortion value from the second moment of transmission distortion computation unit; and

a mode selector unit, the mode selector unit adapted to receive the computed first and second moment of transmission distortion value, the image or video data input, and the feedback from the motion compensation unit.

12. The encoder as recited in claim 11 , wherein one or more of encoder or the rate distortion optimization unit thereof is adapted to reduce transmission distortion at a decoder.

13. An apparatus for encoding an image or video signal according to a method as recited in claim 8 .

14. A system for encoding an image or video signal according to the method as recited in claim 1 .

15. A non-transitory computer readable storage medium containing a set of instructions that causes a computer to perform a method as recited in claim 1 .