KR20070037532A - Intra prediction encoding method and encoder thereof - Google Patents

Abstract

Disclosed are an intra prediction encoding method and a video encoding apparatus that performs intra prediction encoding. An intra prediction encoding method according to the present invention includes: filtering adjacent pixels for predicting an intra 8 × 8 block; Dividing the intra 8 × 8 block into four 4 × 4 blocks; And performing intra 4x4 prediction encoding on the intra 8x8 block by using the filtered neighboring pixels. According to the present invention, when performing the intra 8 × 8 prediction, the size of hardware may be reduced by performing the intra 8 × 8 prediction using the intra 4 × 4 prediction without separately providing an intra 8 × 8 predictor.

Intra Predictive Coding, Filtering, Adjacent Pixels, Encoders, Hardware

Description

An intra prediction encoding method of an image and an encoding apparatus using the method

1 is a diagram for explaining an intra 8x8 prediction mode.

2 is a diagram for explaining an intra 4x4 prediction mode.

3 is a diagram for explaining an intra 16x16 prediction mode.

4 is a diagram for explaining directions of intra 8x8 prediction and 4x4 prediction applied to intra prediction.

5 is a block diagram illustrating a video encoding apparatus according to an embodiment of the present invention.

6A to 6H illustrate a process of performing a 4x4 intra prediction method on 8x8 intra prediction according to an embodiment of the present invention.

7 is a flowchart illustrating an intra prediction encoding method according to an embodiment of the present invention.

The present invention relates to intra prediction encoding, and more particularly, to an intra prediction encoding method, a video encoding apparatus for performing intra prediction encoding, and a computer-readable recording medium having recorded thereon a program for performing intra prediction encoding.

The intra prediction process of H.264 / AVC provides various prediction modes as a method for predictively coding a block in a frame using only information in the same frame. This prediction mode plays a very important role in increasing the compression efficiency of H.264 / AVC. However, there is a problem that the encoder must select which of these modes the compression efficiency is most desirable. In order to select an optimal intra prediction mode, it is common to calculate a rate-distortion cost (RDcost) by performing encoding on all modes and select a mode having the smallest value. However, this method increases the complexity of the encoder.

In addition, intra prediction is a new high profile, and as a result, not only intra 4 × 4 prediction and intra 16 × 16 prediction but also intra 8 × 8 prediction appear, thereby increasing the complexity of the encoder.

In order to solve the above problems, the technical problem to be achieved by the present invention is to use the intra 4x4 prediction method without performing a separate predictor for intra 8x8 prediction when performing intra 8x8 prediction encoding The present invention provides an intra prediction method capable of reducing the size of hardware by performing the same, and an encoding apparatus using the method.

An intra prediction method according to an aspect of the present invention includes: filtering adjacent pixels for predicting an intra 8 × 8 block; Dividing the intra 8 × 8 block into four 4 × 4 blocks; And performing intra 4x4 prediction encoding on the intra 8x8 block by using the filtered neighboring pixels.

Preferably, when performing intra 4x4 prediction encoding for the horizontal-up mode, a prediction block is generated using a predictor in the form of a 4x4 block in the upper left side of the intra 8x8 prediction block.

Preferably, performing the intra 4x4 prediction encoding comprises: generating a prediction block according to each prediction mode using the filtered neighboring pixels for four 4x4 blocks; Calculating a difference block according to a prediction mode by obtaining a difference between a prediction block and a target block to be predicted; Calculating a cost from the difference block according to the prediction mode; Calculating the cost of the intra 8 × 8 prediction block by summing up the four calculated costs; And determining the prediction mode of the differential block having the minimum cost from the costs calculated according to the prediction mode.

According to another aspect of the present invention, a video encoding apparatus filters adjacent pixels for predicting an intra 8 × 8 block, divides the intra 8 × 8 block into four 4 × 4 blocks, and uses intra 8 to filter the adjacent pixels. An intra prediction encoder is configured to perform intra 4 × 4 prediction encoding on a × 8 block.

Hereinafter, with reference to the accompanying drawings will be described in detail a preferred embodiment of the present invention.

The intra prediction process of H.264 / AVC is a method for predictively coding a block in a frame using only information in the same frame. Four 16 × 16 prediction modes and nine 4 × 4 prediction modes for luminance signals are used. And nine recently added 8x8 prediction modes, and there are four 8x8 prediction modes for chrominance signals. A prediction mode of H.264 / AVC will be described with reference to FIGS. 1 to 3.

1 is a diagram for explaining an intra 8x8 prediction mode, and FIG. 2 is a diagram for explaining an intra 4x4 prediction mode.

Intra 8x8 prediction and 4x4 prediction have nine identical prediction modes. In other words, intra 8 × 8 prediction and 4 × 4 prediction include a vertical prediction mode (mode 0), a horizontal prediction mode (mode 1), a DC prediction mode (mode 2), a diagonal down-left prediction mode (mode 3), and a diagonal down. -Have a right prediction mode (mode 4), a vertical-right prediction mode (mode 5), a horizontal-down prediction mode (mode 6), a vertical-left prediction mode (mode 7) and a horizontal-up prediction mode (mode 8) .

The directions of intra 8x8 prediction and 4x4 prediction shown in FIGS. 1 and 2 are the same as shown in FIG. 4. 4 is a diagram illustrating the directions of intra 8x8 prediction and 4x4 prediction applied to intra prediction. The number at the end of the arrow in Figure 4 represents the mode value when the prediction is performed in that direction.

3 is a diagram for explaining an intra 16x16 prediction mode.

The intra 16x16 prediction mode has a vertical prediction mode (mode 0), a horizontal prediction mode (mode 1), a DC prediction mode (mode 2), and a plane prediction mode (mode 3).

5 is a block diagram illustrating an encoding apparatus according to an embodiment of the present invention.

Referring to FIG. 5, the video encoding apparatus includes a transform / quantization unit 110, an inverse quantization / inverse transform unit 131, a deblocking filter unit 133, a picture reconstruction unit 135, a motion compensation prediction unit 137, An intra predictor 139, a motion estimator 150, a subtractor 170, and an entropy coding unit 190 are included.

Image data is input to a video encoding apparatus in units of macroblocks of 16x16 pixels. The transform / quantization unit 110 converts the input macroblocks according to a predetermined method and then quantizes them. A representative image transformation technique is DCT (Discrete Cosing Transform).

The inverse quantization / inverse transform unit 131 receives inverse quantized image data after conversion from the transform / quantization unit 110 and inverse quantizes and inverse transforms. The deblocking filter unit 133 receives inverse quantized and inversely transformed image data from the inverse quantization / inverse transform unit 131 and performs filtering to remove a blocking effect.

The picture reconstruction unit 135 receives the filtered image data from the deblocking filter unit 133 and reconstructs and stores the image in picture units. The picture is an image in a frame unit or an image in a field unit. The picture reconstruction unit 135 includes a buffer (not shown) capable of storing a plurality of pictures. The plurality of pictures stored in the buffer are pictures provided for motion estimation and referred to as reference pictures.

The motion estimator 150 receives at least one reference picture stored in the picture reconstruction unit 135 to perform motion estimation of an input macro block, thereby including motion vector, an index indicating a reference picture, and a block mode. )

The motion compensation predictor 137 corresponds to an input macro block from a reference picture used for motion estimation among a plurality of reference pictures stored in the picture reconstructor 135 according to the motion data input from the motion estimator 150. Extract and output the macro block.

The subtractor 170 receives the macroblock in the reference picture corresponding to the input macroblock from the motion compensation predictor 137 and performs a difference operation with the input macroblock when the input macroblock is predictively encoded between the pictures. Output a signal.

The residual signal output from the subtractor 170 is converted and quantized by the transform / quantization unit 110 and entropy coded by the entropy coding unit 190 to generate an output bit stream.

The intra prediction unit 139 performs intra prediction encoding by using the reconstructed adjacent pixel values in the picture on which the prediction is performed. The reconstructed neighbor pixel value is a value generated by combining the prediction block and the residual signal that has been transformed, quantized, dequantized, and inverse transformed.

Hereinafter, the operation of the intra predictor 139 according to the preferred embodiment of the present invention will be described in detail. The intra predictor 139 filters adjacent pixels to predict an intra 8 × 8 block. Filtering is a process included in the H.264 / AVC standard to improve the image quality during intra 8 × 8 prediction.

Then, the intra predictor 139 divides the intra 8 × 8 block into four 4 × 4 blocks and performs intra 4 × 4 prediction encoding on the intra 8 × 8 block by using the filtered neighboring pixels.

When the intra prediction unit 139 performs intra 4x4 prediction encoding, the prediction block generates prediction blocks according to each of nine prediction modes by using adjacent pixels filtered for four 4x4 blocks. The difference block according to the prediction mode is calculated by obtaining a difference from the original block to be predicted. Then, the intra prediction unit 139 calculates the cost from the differential block according to the prediction mode, calculates the cost for the intra 8 × 8 block by summing the costs for four 4 × 4 blocks, and according to the prediction mode. Determine the prediction mode of the differential block with the minimum cost from the calculated costs.

Here, the cost is a function value indicating the magnitude of the accuracy of prediction encoding and the amount of bits generated. Examples of functions for measuring cost include sum of absolute difference (SAD), sum of absolute transformed difference (SATD), sum of squared difference (SSD), and mean of absolute difference (MAD). For example, the cost using the SAD function is the sum of the absolute value of the difference between the prediction value of each pixel of the current subblock (or macroblock) and the actual pixel value.

Meanwhile, the video decoder 130 that decodes the bit stream generated by the video encoding apparatus described above includes an inverse quantization / inverse transform unit 131, a deblocking filter unit 133, a picture reconstruction unit 135, The motion compensation predictor 137 and the intra predictor 139 are included.

6A through 6H illustrate a process of performing a 4 × 4 intra prediction method on 8 × 8 intra prediction according to an embodiment of the present invention.

FIG. 6A illustrates a process of performing intra prediction on a 8 × 8 intra block using a vertical mode of 4 × 4 blocks. In FIG. 6A, when block 60 is referred to as an intra 4 × 4 vertical mode predictor, intra 8 × 8 prediction is achieved by performing intra 4 × 4 prediction using adjacent pixel values for four 4 × 4 blocks. can do. This is because 8 × 8 intra blocks and 4 × 4 intra blocks have a form of a common prediction block as shown.

Therefore, when adjacent pixels (not shown) immediately above 00, 10, 20, 30, 40, 50, 60, 70 are A, B, C, D, E, F, G, H, the first block In order to generate the predictive block at the upper left, the predicted pixels of 00 to 03 are determined using neighboring pixels A, the predicted pixels of 10 to 13 are determined using neighboring pixels B, and the neighboring pixels C are used. 20 to 23 predicted pixels are determined, and 30 to 33 predicted pixels are determined using the adjacent pixel D. E, F, G, and H are used to generate the prediction block at the upper right corner. A prediction block may be generated by using adjacent pixels A, B, C, and D in the lower left prediction block, and by using adjacent pixels E, F, G, and H in the lower right prediction block.

FIG. 6B uses a 4 × 4 block Horizontal mode for an 8 × 8 intra block, FIG. 6C uses a 4 × 4 block Diagonal Down-Left mode for an 8 × 8 intra block, and FIG. 6D shows an 8 × 8 block. 4 x 4 blocks of Diagonal Down-Right mode are used for 8 intra blocks.

FIG. 6E uses the 4 × 4 block Vertical-Right mode for 8 × 8 intra blocks, FIG. 6F uses the 4 × 4 block Horizontal Down mode for 8 × 8 intra blocks, and FIG. 6G shows 8 × 8 blocks. A vertical-left mode of 4x4 blocks is used for 8 intra blocks, and FIG. 6h uses a 4x4 block horizontal-up mode for 8x8 intra blocks.

In the case of Figs. 6A to 6G, the 4x4 predictors 60, 61, 62, 63, 64, 65 and 66 are used in order to use the 4x4 intra prediction method for 8x8 intra prediction. By differentiating, 8x8 intra prediction may be achieved by performing intra prediction on four 4x4 blocks.

In the case of FIG. 6H, that is, in the case of the horizontal-up mode, the prediction block is formed using the 4 × 4 predictor 67 having the same shape as the 4 × 4 block in the upper left side of the intra 8 × 8 prediction block in the horizontal-up mode. Create The remaining three 4x4 blocks of FIG. 6H differ in form from the 4x4 predictor 67 in the horizontal up mode, but similarly to the cases of FIGS. 6A to 6G, prediction blocks are generated. Although not shown, in the DC mode of the intra 8 × 8 prediction, the prediction block may be generated by performing intra 4 × 4 prediction four times and generating a prediction block and dividing by 4 for the pixel value of the prediction block. will be.

7 is a flowchart illustrating an intra prediction encoding method according to an embodiment of the present invention.

First, adjacent pixels for predicting an intra 8 × 8 block are filtered (S 110).

The intra 8 × 8 block is divided into four 4 × 4 blocks (S 120). Intra 4 × 4 prediction encoding is performed on the intra 8 × 8 block by using the filtered neighboring pixels (S130).

When performing intra 4x4 prediction encoding, a prediction block is generated according to each prediction mode by using adjacent pixels filtered for four 4x4 blocks, and the prediction block and the original block to be predicted Calculate the difference block according to the prediction mode by obtaining the difference from. Then, the cost is calculated from the differential block according to the prediction mode, and the four costs are summed to calculate the cost of the intra 8x8 block.

The prediction mode of the differential block having the minimum cost is determined from the calculated costs according to the prediction mode.

The intra prediction method according to the present invention can be embodied as computer readable codes on a computer readable recording medium. Computer-readable recording media include all kinds of recording devices that store data that can be read by a computer system. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical disk, and the like, and may also include those implemented in the form of carrier waves (eg, transmission over the Internet). do. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.

The above description is only one embodiment of the present invention, and those skilled in the art may implement the present invention in a modified form without departing from the essential characteristics of the present invention. Therefore, the scope of the present invention should not be limited to the above-described examples, but should be construed to include various embodiments within the scope equivalent to those described in the claims.

As described above, according to the present invention, when performing the intra 8 × 8 prediction, the intra 8 × 8 prediction is performed using the intra 4 × 4 prediction to reduce the size of the hardware without separately providing an intra 8 × 8 predictor. Can be.

Claims (5)

Filtering adjacent pixels to predict intra 8 × 8 blocks; Dividing the intra 8 × 8 block into four 4 × 4 blocks; And And performing intra 4x4 predictive encoding on the intra 8x8 block by using the filtered neighboring pixels. The method of claim 1, When performing intra 4 × 4 prediction coding for the horizontal-up mode, An intra prediction method comprising generating a prediction block using a 4 × 4 block type predictor in the upper left side of an intra 8 × 8 prediction block. The method of claim 1, wherein the performing of intra 4 × 4 prediction encoding is performed. For the four 4 × 4 blocks, Generating a prediction block according to each prediction mode using the filtered neighboring pixels; Calculating a difference block according to the prediction mode by obtaining a difference between the prediction block and a target block to be predicted; Calculating a cost from the difference block according to the prediction mode; Calculating the cost of an intra 8 × 8 prediction block by summing up the four calculated costs; And And determining a prediction mode of a differential block having a minimum cost from the calculated costs according to the prediction mode. Filter adjacent pixels to predict intra 8 × 8 blocks, And an intra prediction encoder for dividing the intra 8 × 8 block into four 4 × 4 blocks and performing intra 4 × 4 prediction encoding on the intra 8 × 8 block by using the filtered neighboring pixels. Video encoding device. A computer-readable recording medium having recorded thereon a program for implementing the intra prediction coding method of claim 1.

Images