KR101811774B1 - Apparatus and method for processing graphics - Google Patents

Abstract

The present invention relates to a graphics processing apparatus and method. The graphic processing method receives image data, selects whether to compress the image data by a lossy compression method or a lossless compression method, and compresses the image data by the selected compression method. Then, the compressed data is stored in the frame buffer allocated in the memory. According to the present invention, a process of compressing or losslessly compressing or lossily compressing or losing data is selectively operated according to the nature and purpose of the image data, thereby compressing data input / output to / from the frame buffer disposed in the off- Can be improved.

Description

[0001] Apparatus and method for processing graphics [

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a graphics processing apparatus and method, and more particularly, to a graphics processing apparatus and method capable of compressing input and output data to and from a frame buffer by selectively using a lossy compression method and a lossless compression method.

In general, the bandwidth for storing image data in the frame buffer or for fetching from the frame buffer is limited. However, the necessity of high-quality image processing such as QHD (Quad High Definition) and UHD (Ultra High Definition) is increasing, and in order to solve such a bandwidth problem, Frame Buffer Compression ) Method has been proposed.

Data compression refers to a technique for efficiently recording data in less storage space, or a practical application of the technique. Data compression consists of an encoding process that largely transforms the data to a smaller size and a decoding process that recalls the stored data and restores the original data form. At this time, the ratio of the data size before encoding to the data size after encoding is called compression rate.

Depending on the type of data compression technique, lossless compression, which can be decoded as it is without changing the contents of the data, and lossy compression, which achieves a higher compression rate but sacrifices some of the detailed detail of the decoded data lossy compression. Representative lossless compression schemes include iterative length coding, Huffman coding, and arithmetic coding. Since the lossy compression method increases the compression rate by reversing the characteristics of human sensory organs, different algorithms are used depending on the type of data such as voice, still picture, and moving picture.

However, most frame buffer compression schemes for compressing data input and output to and from the frame buffer use a lossy compression scheme in order to obtain high compression efficiency. However, there is a problem in that lossy compression not only degrades the quality of the image but also makes it difficult to restore original data when necessary. Also, when a lossless compression scheme is used to prevent this, a remarkable bandwidth reduction effect may not be obtained.

Accordingly, when a fixed compression method is used irrespective of the nature of the processed image data, efficient processing can not be performed. Therefore, a method of losslessly compressing or loss-compressing image data is selectively performed according to the operating environment, There is a need to do.

Accordingly, it is an object of the present invention to provide a graphics processing apparatus and method capable of selectively compressing a lossy compression method and a lossless compression method for data input to and output from a frame buffer.

According to another aspect of the present invention, there is provided a graphic processing method comprising the steps of: receiving image data; selecting a compression method of the lossy compression method or a lossless compression method; Compressing the data, and storing the compressed data in a frame buffer disposed in the memory.

The method may further include extracting compressed data stored in the frame buffer, and decompressing the compressed data to generate decompressed data.

According to another aspect of the present invention, there is provided a graphics processing apparatus including an encoding unit for compressing inputted image data by either a lossy compression method or a lossless compression method, A decoding unit for decoding the compressed data extracted from the frame buffer, and a control logic unit for controlling operations of the respective units.

According to another aspect of the present invention, there is provided a computer-readable recording medium having recorded thereon a program for causing a processor to execute the graphic processing method.

According to the present invention, there is provided a 3D graphics processor including the graphic processing device.

According to the present invention, when inputting / outputting image data to / from a frame buffer arranged in an off-chip memory, a lossy compression method and a lossless compression method can be selectively used for compression. This reduces memory storage space and reduces access to off-chip memory, which consumes a lot of power, thereby reducing memory bandwidth and lowering power usage, thereby beneficially improving overall performance.

1 is a block diagram illustrating a configuration of a graphics processing apparatus according to an exemplary embodiment of the present invention;
FIG. 2 is a block diagram showing an example of the configuration of the graphic processing unit shown in FIG. 1, and
FIG. 3 and FIG. 4 are flowcharts provided in a description of a graphic processing method according to an embodiment of the present invention.

Hereinafter, the present invention will be described in detail with reference to the drawings.

1 is a block diagram illustrating a configuration of a graphics processing apparatus according to an embodiment of the present invention.

Referring to FIG. 1, the graphic processing apparatus 100 may include an image input unit 110, a memory 120, a display device 130, and a graphics processing unit 140. When such components are implemented in practical applications, two or more components may be combined into one component, or one component may be divided into two or more components as necessary.

The image input unit 110 is for inputting image data, and may include a device for receiving a camera or other image data. The image input unit 110 can process image frames such as still images and moving images obtained by an image sensor or the like.

A frame buffer is disposed in the memory 120 to store image data. In addition, data input or output through the graphic processor 140 may be stored.

The display device 130 displays the image data processed by the graphic processing unit 140.

The graphic processing unit 140 may store the image data input through the image input unit 110 in a frame buffer disposed in the memory 120 and fetch data stored in the frame buffer and display the data on the display device 130. [

In such a configuration, when storing data in the frame buffer arranged in the memory 120, the graphic processing unit 140 may store the compressed data using a lossless compression scheme or a lossy compression scheme. That is, the graphic processing unit 140 may compress the image data using the lossy compression method or the lossless compression method, and then store the compressed image data in the memory 120. The compressed data stored in the memory 120 is restored by the graphic processing unit 140 and used.

Such compression storage can advantageously improve overall performance by reducing the storage space of memory 120 and by preventing bottlenecks by reducing the bandwidth load for reading or writing on the bus.

2 is a block diagram showing an example of the configuration of the graphics processing unit shown in FIG.

2, the graphic controller 140 includes an encoding unit 141, a decoding unit 142, a display sub-unit 143, an interface unit 144, a memory control unit 145, a control logic unit 146, And an internal bus and connection circuitry 147.

The encoding unit 141 compresses the input image data by either a lossy compression method or a lossless compression method. The encoding unit 141 may generate reconstruction information for reconstructing original data when the image data is compressed by the lossy compression method.

The decoding unit 142 decompresses the compressed data fetched from the memory 120. The decoding unit 142 may restore original data before compression using the reconstruction information generated and stored by the encoding unit 141 upon decompression of the compressed data due to the lossy compression.

The display sub-unit 143 reads the decompressed data from the decoding unit 142 and displays it on the display device 130. [

The interface unit 144 is connected to the image input unit 110 so that it can input and output image data.

The memory control unit 145 controls access to the memory 120 and stores the compressed data in the encoding unit 141 in the frame buffer disposed in the memory 120 and can extract the compressed data stored in the frame buffer have,

The control logic unit 146 typically controls the operation of each of the units to control the overall operation of the graphics processing unit 140.

The internal bus and connection circuitry 147 includes an encoding section 141, a decoding section 142, a display sub-section 143, an interface section 144, a memory control section 145, and a control logic section 146, .

The graphic processing unit 140 compresses the image data input through the image input unit 110 using one of a lossy compression method and a lossless compression method, Lt; / RTI > That is, compression is performed using a lossless compression method for image data which must be losslessly compressed, but compression can be performed using lossy compression for image data that can be lossy compressed. .

In mobile processors, the bus performance of data is becoming an immediate problem due to the constrained environment, while the semiconductor process is becoming finer and the performance of the processor can be further increased. In this case, compression technology is used. By compressing the data in order to increase the efficiency of the data, the memory can be expected to be wider. have. There is a loss of data compression and decompression, but this is a problem that can be solved if chip performance improves.

FIG. 3 and FIG. 4 are flowcharts provided in a description of a graphic processing method according to the present invention.

Referring to FIG. 3, when image data is input through the image input unit 110 (S300), the graphic processing unit 140 uses either a lossy compression method or a lossless compression method depending on a preset reference, The data compression method is selected (S305).

If the lossy compression method is selected as the data compression method (S310), the graphic processing unit 140 compresses the image data using the lossy compression method (S315). At this time, restoration information for restoring original data may be additionally generated (S320). The restoration information is used to restore original data before compression even if the compressed data is stored as lossy compression, if necessary, using the restored information.

The graphic processing unit 140 stores the compressed data and reconstruction information in the memory 110 (S325).

If the lossless compression method is selected as the data compression method (S330), the graphic processing unit 140 compresses the image data using the lossless compression method (S335), and stores the compressed data in the memory 110 (S340).

Referring to FIG. 4, if the display of the image data or other image data stored in the frame buffer is required, the graphic processor 140 fetches the compressed data stored in the frame buffer of the memory 120 (S400).

If the retrieved data is data compressed by the lossy compression method (S405), if restoration of the original data before compression is required (S410), the graphic processing unit 140 restores the original data before compression using the restoration information S415). However, when the restoration of the original data is not necessary, the compression data is decompressed without using the restoration information.

The decompressed and restored image data can be displayed on the display device 130 (S420).

If the extracted data is data compressed and stored in the lossless compression scheme (S430), the graphic processing unit 140 decompresses the compressed data in accordance with the lossless compression scheme (S435). Similarly, decompressed and restored image data can be displayed on the display device 130 (S420).

In this manner, data input to and output from the frame buffer can be compressed using either a lossy compression method or a lossless compression method. In addition, even if lossy compression is used, the original data before compression can be restored using the reconstruction information.

Such compression-based data storage plays an important role in reducing both traffic and power consumption for off-chip memory. In other words, it is very effective in reducing memory bandwidth and power consumption by avoiding power-intensive off-chip memory accesses.

On the other hand, the lossy compression method and the lossless compression method may be selected for each frame or layer screen, or may be selected for each tile. Generally, in a tile-based graphics processing device, rendering is performed for each tile. In the tile-based method, the screen is divided into several tiles and the tiles are rendered one on-chip tile buffer one at a time.

Also, in the case of data compressed by the lossy compression method, the process of restoring data using the reconstruction information may be configured to be applied step by step according to the resolution. For example, HD image data can be generated using the reconstruction information in the SD image quality compressed image data, and UHD image data can be generated using the reconstruction information in the HD image quality compressed image data.

And, the contents of the present invention are not limited to hardware or software use, but are applicable to any other computing or processing environment. But may be embodied in hardware, software, or a combination of hardware and software described in the present invention. The present invention may be implemented using circuitry. That is, it can be implemented as one or more programmable logic circuits, that is, an application specific integrated circuit (ASIC) or a logic circuit (AND, OR NAND gates) or a processing device (for example, a microprocessor, a controller).

The present invention can be implemented as a computer program on a programmable computer. Such a computer may include a processor, a storage device, an input device, and an output device. In order to implement the contents described in the present invention, the program code may be input to a mouse or a keyboard input device. These programs can be implemented in high-level languages or object-oriented languages. It can also be implemented as a computer system implemented in assembly or machine code.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, It should be understood that various modifications may be made by those skilled in the art without departing from the spirit and scope of the present invention.

110: image input unit 120: memory
130: Display device 140: Graphics processor

Claims (13)

Selecting the lossy compression method or the lossless compression method to compress the input image data, and compressing the image data according to the selected compression method;
Storing the compressed data in a frame buffer disposed in a memory and generating reconstruction information for restoring original data when the lossy compression method is selected and storing the reconstructed information in the memory;
Fetching compressed data stored in the frame buffer;
Decompressing the extracted compressed data to generate decompressed data when the extracted compressed data is data compressed by the lossless compression method;
If the extracted compressed data is data compressed by the lossy compression method, if restoration of the original data is required, the decompressed information is applied step by step according to a resolution set at decompression of the extracted compressed data, Generating decompressed data by decompressing the extracted compressed data without applying the decompressed information if decompression of the original data is not required; And
And displaying the decompressed data on a display device. delete delete The method according to claim 1,
Wherein the selection and compression of the compression scheme can be selected in frame units, layer screen units, or tile units. delete delete A processor-readable recording medium having recorded thereon a program for causing a processor to execute the graphic processing method according to claim 1 or claim 4. An encoding unit for compressing input image data by a lossy compression method or a lossless compression method;
A memory controller for storing the compressed data in a frame buffer disposed in a memory and fetching compressed data stored in the frame buffer;
A decoding unit for decompressing compressed data extracted from the frame buffer;
A display sub-unit for displaying decompressed data in the decoding unit on a display device; And
And a control logic unit for controlling operations of the encoding unit, the memory control unit, the decoding unit, and the display sub-unit,
The control logic unit,
Compresses the image data in a compression mode selected from a lossy compression mode and a lossless compression mode, and generates reconstruction information for restoring original data when the lossy compression mode is selected, and controls the encoding unit to be stored in the memory,
When the compressed data fetched from the frame buffer is data compressed by the lossless compression method, decompressing the fetched compressed data,
When the compressed data extracted from the frame buffer is data compressed by the lossy compression method and the original data needs to be decompressed, the decompression information is applied stepwise according to the resolution set at decompression of the extracted compressed data Decompression,
And controls the decoding unit to decompress the extracted compressed data without applying the restoration information if restoration of the original data is not necessary. 9. The method of claim 8,
Wherein the control logic unit can select and compress the compression scheme by a frame unit, a layer screen unit, or a tile unit. delete delete delete 10. A 3D graphics processor comprising the graphics processing device of claim 8 or 9.

Images