System and method of error control for video coding

1. An encoder in an error control system for video coding, comprising a base layer encoding apparatus and at least one enhancement layer encoding apparatus, the encoder further comprising:

an encoder drift control apparatus, configured to adjust a value of an encoder leaky factor according to local information of the base layer encoding apparatus, wherein the local information is number of quantized transformation non-zero coefficients of a current block in the base layer;

and the enhancement layer encoding apparatus is configured to attenuate a motion-compensated signal of the current block in the enhancement layer according to the adjusted value of the encoder leaky factor to control an error drift;

wherein the encoder drift control apparatus is configured to adjust the value of the encoder leaky factor of the current block, alphaCurr, according to the following formula:

if 0 =nCTrs , then alphaCurr=alphaMax*(1 −sF*nCB ); else alphaCurr is zero;

wherein nCB is a number of quantized transformation non-coefficients of a current block in the base layer, nCTrs is a predetermined threshold for the quantized transformation non-coefficients, alphaMax is a maximum leaky factor, and sF is a scale;

wherein the maximum leaky factor alphaMax equals to 1; or the maximum leaky factor alphaMax equals to (max_cliff ref scale for zero_base_block-I-1)/32; and wherein a syntax element max_cliff ref scale_for_zero_base_block, with a range of 0-31, is set as a maximum scale factor for scaling a differential reference signal according to a present SVC standard.

2. The encoder according to claim 1 , wherein the base layer encoding apparatus comprises a first encoder motion compensation apparatus; and wherein the enhancement layer encoding apparatus comprises a second encoder motion compensation apparatus, a third encoder subtraction apparatus and an encoder leaky factor apparatus;

wherein the third encoder subtraction apparatus is configured to combine the motion-compensated signal from the first encoder motion compensation apparatus and the motion-compensated signal from the second encoder motion compensation apparatus; and

wherein the encoder leaky factor apparatus is configured to attenuate the combined signal from the third encoder subtraction apparatus according to the value of the encoder leaky factor, to control the error drift.

3. An decoder in an error control system for video coding, comprising a base layer decoding apparatus and at least one enhancement layer decoding apparatus, the decoder further comprising:

a decoder drift control apparatus, configured to adjust a value of a decoder leaky factor according to local information of the base layer decoding apparatus, wherein the local information is number of quantized transformation non-zero coefficients of a current block in the base layer; and the enhancement layer decoding apparatus is configured to attenuate a motion-compensated signal of the current block in the enhancement layer according to the value of the decoder leaky factor to control an error drift;

wherein the decoder drift control apparatus ( 240 ) is configured to adjust a value of an encoder leaky factor of the current block, alphaCurr, according to the following formula:

if 0 <nCB<=nCTrs , then alphaCurr=alphaMax*(1 −sF*nCB ); else alphaCurr is zero;

wherein nCB is thea number of quantized transformation non-coefficients of a current block in the base layer, nCTrs is a predetermined threshold for the quantized transformation non-coefficients, alphaMax is a maximum leaky factor, and sF is a scale factor;

wherein the maximum leaky factor alphaMax equals to 1; or the maximum leaky factor alphaMax equals to (max diff_ref_scale_for_zero_base_block+1)/32; and

wherein a Syntax element max_diff_ref_scale_for_zero_base_block, with a range of 0-31, is set as a maximum scale factor for scaling a differential reference signal according to a present SVC standard.

4. The decoder according to claim 3 , wherein the base layer decoding apparatus comprises a first decoder motion compensation apparatus; and wherein the enhancement layer decoding apparatus comprises a second decoder motion compensation apparatus, a first decoder subtraction apparatus and an decoder leaky factor apparatus;

wherein the first decoder subtraction apparatus is configured to combine the motion-compensated signal from the first decoder motion compensation apparatus and the motion-compensated signal from the second decoder motion compensation apparatus; and

wherein the decoder leaky factor apparatus is configured to attenuate the combined signal from the first decoder subtraction apparatus according to the value of the decoder leaky factor, to control the error drift.

5. The decoder according to claim 3 , wherein, in a process of adjusting the value of the decoder leaky factor by the decoder drift control apparatus, the decoder drift control apparatus selects the local information of the base layer decoding apparatus according to identification information which is transmitted from an encoding end and which is used for identifying a process of selecting the local information by the encoding end, or the decoder drift control apparatus utilizes a received leaky factor which is adjusted by the decoder drift control apparatus as the decoder leaky factor.

6. An error control method for video coding, comprising:

adjusting a value of an encoder/decoder leaky factor according to a local information of the encoder/decoder,

wherein the local information is number of quantized transformation non-zero coefficients of a current block in a base layer; and

attenuating a motion-compensated signal of the current block in an enhancement layer according to the value of the encoder/decoder leaky factor, to control an error drift;

wherein the adjusting the value of a leaky factor according to local information comprises:

adjusting the value of the encoder leaky factor of the current block, alphaCurr, according to the following formula:

if 0 <nCB<=nCTrs , then alphaCurr=alphaMax*(1 −sF*nCB ); else alphaCurr is zero;

wherein nCB is a number of quantized transformation non-coefficients of a current block in the base layer, nCTrs is a predetermined threshold for the quantized transformation non-coefficients, alphaMax is a maximum leaky factor, and sF is a scale factor;

wherein the maximum leaky factor alphaMax equals to 1; or the maximum leaky factor alphaMax equals to (max_diff_ref_scale_for_zero_base_block+1)/32; and

wherein a Syntax element max_diff_ref_scale_for_zero_base_block, with a range of 0-31, is set as a maximum scale factor for scaling a differential reference signal according to a present SVC standard.

7. The method according to claim 6 , further comprising:

transmitting an identification information which is used for identifying a process of selecting the local information from an encoding end to a decoding end.