Re-sampling with phase offset adjustment for luma and chroma to signal adaptive filters in scalable video coding
View Patent ↗A sampling filter process is provided for scalable video coding. The process provides for re-sampling using video data obtained from an encoder or decoder process of a base layer (BL) in a multi-layer system using adaptive phase shifting to improve quality in Scalable High efficiency Video Coding (SHVC). In order to compensate for phase offsets introduced by downsampling an appropriate phase offset adjustment is made for upsampling in SHVC with an appropriate offset included for proper luma/chroma color space positions. In one approach, an adaptive filter is signaled with its phase offset adjusted to account for the luma/chroma offset.
1. A system for scalable video coding comprising:
a first coding layer for coding video with a base resolution;
a second coding layer for coding video with an enhanced resolution having a higher resolution than the base resolution;
an upsampling unit receiving sampling signals from the first coding layer and providing an output signal to the second coding layer after an upsampling process, wherein the upsampling unit output signal enables more efficient coding in the second coding layer, wherein the upsampling unit comprises:
a first module formed by software executable in a processor for selecting input samples of the video signal in the first coding layer;
a second module formed by code executable in the processor providing selection of one adaptive filter of a plurality of adaptive filters for processing the selected input samples; and
a third module including the plurality of adaptive filters, the third module formed by code executable in the processor and providing the output signal from the upsampling unit, wherein the first coding layer is downsampled from the second coding layer, and wherein at least one of a luma phase offset and a chroma phase offset in downsampling is used in determining the adaptive filter,
wherein both the luma phase offset and the chroma phase offset in downsampling is determined and used in determining the adaptive filter,
wherein the luma phase offset and the chroma phase offset are specified for both row and column filtering,
wherein for each filter in the plurality of filters a determination is made of the number of filters in a row from num_phase_offsets_minus1[0] and in a column from num_phase_offsets_minus1[1],
wherein a row and column luma filter is specified differentially using a reference luma filter specified by ref_luma_filter_indx, and using filter coefficient delta values specified by delta_luma_filter_coef, and
wherein a row and column luma filter is specified differentially using a reference chroma filter specified by ref_chroma_filter_indx, and using filter coefficient delta values specified by delta_chroma_filter_coef.
2. The system of claim 1 , wherein a sample location xRef and yRef for the first module with a phase xPhase and yPhase for use in the second module is determined using variables xRefPhase and yRefPhase in the following equations:
x Refphase=( xP *Pic WRL *(num_phase_offsets_minus1[0]+1))/ScaledW
y Refphase=( yP *Pic HRL *(num_phase_offsets_minus1[1]+1))/ScaledH
x Ref=( x Refphase/(num_phase_offsets_minus1[0]+1))
x Phase=( x Refphase- x Ref*(num_phase_offsets_minus1[0]+1))
y Ref=( y Refphase/(num_phase_offsets_minus1[1]+1))
y Phase=( y Refphase- y Ref*(num_phase_offsets_minus1[1]+1))
wherein xP and yP are a sample location relative to the top left sample, PicWRL is the picture width reference layer, PicHRL is the picture height reference layer, and ScaledW and ScaledH are the scalable width and height in the second coding layer.
3. The system of claim 1 , wherein it is signaled from an encoder to a decoder using a flag in the second coding layer whether the first coding layer data is deblocked data.
4. The system of claim 1 , wherein when data has been deblocked it is signaled from an encoder to a decoder using a flag in the second coding layer whether the first coding layer data has been processed with SAO.
5. The system of claim 1 , wherein the selection of a filter from the plurality of filters in the third module is controlled by selection from the second module based on the first phase offset determination.
6. A method for scalable video coding comprising: receiving sampling signals from the first coding layer and providing an output signal for an upsampling process to a second coding layer that codes video with an enhanced resolution having a higher resolution than the base resolution;
selecting input samples of the video signal in the first coding layer for coding video with a base resolution; selecting from a plurality of filters for processing the selected input samples; and
filtering the selected input samples with the selected filter, wherein an adaptive phase offset is generated for the selected filter to be used in an upsampling process to provide the output signal for the upsampling process, and wherein at least one of luma phase offset and chroma phase offset is determined and included in the adaptive phase offset of the selected filter,
wherein both the luma phase offset the chroma phase offset in downsampling is determined and used in providing the adaptive phase offset of the selected filter,
wherein the luma phase offset and the chroma phase offset are specified for both row and column filtering,
wherein for each filter in the plurality of filters a determination is made of the number of filters in a row from num_phase_offsets_minus1[0] and in a column from num_phase_offsets_minus1[1],
wherein a row and column luma filter is specified differentially using a reference luma filter specified by ref_luma_filter_indx, and using filter coefficient delta values specified by delta_luma_filter_coef, and
wherein a row and column luma filter is specified differentially using a reference chroma filter specified by ref_chroma_filter_indx, and using filter coefficient delta values specified by delta_chroma_filter_coef.
7. The method of claim 6 , wherein a sample location xRef and yRef for the first module with a phase xPhase and yPhase for use in the second module is determined using variables xRefPhase and yRefPhase in the following equations:
x Refphase=( xP *Pic WRL *(num_phase_offsets_minus1[0]+1))/ScaledW
y Refphase=( yP *Pic HRL *(num_phase_offsets_minus1[1]+1))/ScaledH
x Ref=( x Refphase/(num_phase_offsets_minus1[0]+1))
x Phase=( x Refphase- x Ref*(num_phase_offsets_minus1[0]+1))
y Ref=( y Refphase/(num_phase_offsets_minus1[1]+1))
y Phase=( y Refphase- y Ref*(num_phase_offsets_minus1[1]+1))
wherein xP and yP are a sample location relative to the top left sample, PicWRL is the picture width reference layer, PicHRL is the picture height reference layer, and ScaledW and ScaledH are the scalable width and height in the second coding layer.
8. The method of claim 6 , wherein it is signaled from an encoder to a decoder using a flag in the second coding layer whether the first coding layer data is deblocked data.
9. The method of claim 6 , wherein when data has been deblocked it is signaled from an encoder to a decoder using a flag in the second coding layer whether the first coding layer data has been processed with SAO.