Intra-Prediction Estimation Using Approximate Reconstructed Samples

Abstract

A method for intra-prediction estimation is provided that includes determining a best intra-prediction mode for a block of samples, wherein at least some of the neighboring samples used for intra-prediction estimation include approximate reconstructed samples, applying approximate reconstruction to the block of samples using the best intra-prediction mode to generate a block of approximate reconstructed samples, and storing the block of approximate reconstructed samples for use in intra-prediction estimation of other blocks of samples.

What is claimed is: 
     
         1 . A method for intra-prediction estimation, the method comprising:
 determining a best intra-prediction mode for a block of samples, wherein at least some of the neighboring samples used for intra-prediction estimation include approximate reconstructed samples;   applying approximate reconstruction to the block of samples using the best intra-prediction mode to generate a block of approximate reconstructed samples; and   storing the block of approximate reconstructed samples for use in intra-prediction estimation of other blocks of samples.   
     
     
         2 . The method of  claim 1 , wherein applying approximate reconstruction comprises applying approximate reconstruction to the block of samples when the intra-prediction mode for the block is better than intra-prediction modes selected for child blocks of the block of samples. 
     
     
         3 . The method of  claim 1 , wherein applying approximate reconstruction comprises:
 applying a simplified transform to a block of residual sample values of the block of samples to generate a block of transform coefficients;   applying simplified quantization to the block of transform coefficients to generate a block of quantized transform coefficients;   applying inverse quantization to the block of quantized transform coefficients to recover the block of transform coefficients;   applying an inverse of the simplified transform to the recovered block of transform coefficients to generate a block of approximate residual sample values; and   adding the block of approximate residual sample values to a corresponding block of predicted samples to generate the block of approximate reconstructed samples.   
     
     
         4 . The method of  claim 3 , wherein the simplified transform is one selected from a group consisting of a transform derived from a transform specified in a video coding standard and a Hadamard transform. 
     
     
         5 . The method of  claim 4 , wherein the transform derived from the transform specified in the video coding standard is one selected from a group consisting of a 4×4 simplified transform and an 8×8 simplified transform, and the Hadamard transform is one selected from a group consisting of a 4×4 Hadamard transform and an 8×8 Hadamard transform. 
     
     
         6 . The method of  claim 3 , wherein the simplified transform is a smaller transform than a transform used for encoding the block of samples. 
     
     
         7 . The method of  claim 3 , wherein the simplified quantization comprises one selected from a group consisting of quantization in which precision is reduced from quantization specified in a video coding standard, quantization in which mathematical operations specified for quantization in a video coding standard are replaced by left and right shift operations, and quantization in which precision is reduced from quantization specified in a video coding standard. 
     
     
         8 . An apparatus configured to perform intra-prediction estimation, the apparatus comprising:
 means for determining a best intra-prediction mode for a block of samples, wherein at least some of the neighboring samples used for intra-prediction estimation include approximate reconstructed samples;   means for applying approximate reconstruction to the block of samples using the best intra-prediction mode to generate a block of approximate reconstructed samples; and   means for storing the block of approximate reconstructed samples for use in intra-prediction estimation of other blocks of samples.   
     
     
         9 . The apparatus of  claim 8 , wherein the means for applying approximate reconstruction comprises means for applying approximate reconstruction to the block of samples when the intra-prediction mode for the block is better than intra-prediction modes selected for child blocks of the block of samples 
     
     
         10 . The apparatus of  claim 8 , wherein the means for applying approximate reconstruction comprises:
 means for applying a simplified transform to a block of residual sample values of the block of samples to generate a block of transform coefficients;   means for applying simplified quantization to the block of transform coefficients to generate a block of quantized transform coefficients;   means for applying inverse quantization to the block of quantized transform coefficients to recover the block of transform coefficients; and   means for applying an inverse of the simplified transform to the recovered block of transform coefficients to generate a block of approximate residual sample values; and   means for adding the block of approximate residual sample values to a corresponding block of predicted samples to generate the block of approximate reconstructed samples.   
     
     
         11 . The apparatus of  claim 10 , wherein the simplified transform is one selected from a group consisting of a transform derived from a transform specified in a video coding standard and a Hadamard transform. 
     
     
         12 . The apparatus of  claim 11 , wherein the transform derived from the transform specified in the video coding standard is one selected from a group consisting of a 4×4 simplified transform and an 8×8 simplified transform, and the Hadamard transform is one selected from a group consisting of a 4×4 Hadamard transform and an 8×8 Hadamard transform. 
     
     
         13 . The apparatus of  claim 10 , wherein the simplified transform is a smaller transform than a transform used for encoding the block of samples. 
     
     
         14 . The apparatus of  claim 10 , wherein the simplified quantization comprises one selected from a group consisting of quantization in which precision is reduced from quantization specified in a video coding standard, quantization in which mathematical operations specified for quantization in a video coding standard are replaced by left and right shift operations, and quantization in which precision is reduced from quantization specified in a video coding standard. 
     
     
         15 . A non-transitory computer readable medium storing software instruction that, when executed by at least one processor, cause the execution of a method for intra-prediction estimation, the method comprising:
 determining a best intra-prediction mode for a block of samples, wherein at least some of the neighboring samples used for intra-prediction estimation include approximate reconstructed samples;   applying approximate reconstruction to the block of samples using the best intra-prediction mode to generate a block of approximate reconstructed samples; and   storing the block of approximate reconstructed samples for use in intra-prediction estimation of other blocks of samples.   
     
     
         16 . The computer readable medium of  claim 15 , wherein applying approximate reconstruction comprises applying approximate reconstruction to the block of samples when the intra-prediction mode for the block is better than intra-prediction modes selected for child blocks of the block of samples. 
     
     
         17 . The computer readable medium of  claim 15 , wherein applying approximate reconstruction comprises:
 applying a simplified transform to a block of residual sample values of the block of samples to generate a block of transform coefficients;   applying simplified quantization to the block of transform coefficients to generate a block of quantized transform coefficients;   applying inverse quantization to the block of quantized transform coefficients to recover the block of transform coefficients;   applying an inverse of the simplified transform to the recovered block of transform coefficients to generate a block of approximate residual sample values; and   adding the block of approximate residual sample values to a corresponding block of predicted samples to generate the block of approximate reconstructed samples.   
     
     
         18 . The computer readable medium of  claim 17 , wherein the simplified transform is one selected from a group consisting of a transform derived from a transform specified in a video coding standard and a Hadamard transform. 
     
     
         19 . The computer readable medium of  claim 17 , wherein the simplified transform is a smaller transform than a transform used for encoding the block of samples. 
     
     
         20 . The computer readable medium of  claim 17 , wherein the simplified quantization comprises one selected from a group consisting of quantization in which precision is reduced from quantization specified in a video coding standard, quantization in which mathematical operations specified for quantization in a video coding standard are replaced by left and right shift operations, and quantization in which precision is reduced from quantization specified in a video coding standard.