JP3998634B2 - Moving picture encoding apparatus and moving picture decoding apparatus - Google Patents

Description

  The present invention relates to a moving image encoding apparatus that encodes and transmits information such as an image signal, and a moving image decoding apparatus that decodes encoded information.

  In recent years, with the spread of ISDN (Integrated Services Digital Network), an image communication service has been realized as a new communication service. Examples are videophones and videoconferencing systems. In addition, with the development of wireless transmission networks represented by PHS and FPLMTS, demands for further advancement, diversification, and portability of services are rapidly increasing.

  In general, when transmitting image information, such as a videophone or a video conference system, the amount of image information is enormous. On the other hand, the amount of image information to be transmitted depends on the line speed and cost of the line used for transmission. It is necessary to compress and encode and transmit with a reduced amount of information.

  As coding methods for compressing image information, JPEG (Joint Photographic Coding Experts Group) is used as a still image coding method, and H.264 is used as a moving image coding method. 261, MPEG (Moving Picture Coding Expert Group) 1 and MPEG 2 have already been internationally standardized as moving picture coding systems for storage. Furthermore, MPEG4 standardization activities are underway as an encoding method at an extremely low bit rate of 64 kbps or less.

  MPEG4 can flexibly handle a wide variety of applications, and can be encoded by an optimum method for each application. There are many encoder tools (transformers, quantizers, inverse transformers, inverse quantizers, etc.) instead of coding according to an algorithm like the H.261, MPEG1, and MPEG2 coding systems. Therefore, it is necessary that the encoding is performed by an appropriate combination thereof.

  FIG. This is a data string of encoded output information encoded by H.261, and FIG. 5B is a data string of encoded output information encoded by an encoding method with a flexible algorithm. In the case of FIG. 11B, since the combination of each tool of the encoder can be freely selected, it is necessary to transmit information relating to the tool indicating which tool was used for encoding together with the encoded output information. There is.

  In FIG. 11, motion compensation tool A, inverse transform tool B, motion compensation tool C, inverse transform tool D, and quantization tool E are tool information, followed by motion vector information 101, transform coefficient 102, motion vector information 103, transform The coefficient 104 and the quantization step 105 are processed data.

  FIG. FIG. 2 is a block diagram illustrating an example of a conventional encoding / decoding device according to H.261. This encoding / decoding device includes a control unit 106 that controls the entire device, an encoder 107 that performs encoding according to H261, The decoder 108 includes a decoder 108 that decodes information encoded by the H.261, and a tool storage unit 109 that includes a memory that stores tool information.

  Considering the configuration for realizing these encoding algorithms, there are a method realized by mounting dedicated hardware and software, and a method realized by executing appropriate software on a general-purpose arithmetic unit. .

  FIG. 13 is a block diagram of the encoder 107. The encoder 107 includes an encoding control unit 111 that performs encoding control, a conversion unit 112 that performs DCT conversion, a quantization unit 113 that performs quantization of the coefficient converted by the conversion unit, and an inverse of the quantized coefficient. Inverse quantization unit 114 that performs quantization, inverse transform unit 115 that performs inverse DCT transform, memory 116 having a variable delay function for motion compensation used in motion compensation inter-frame prediction, in a loop that can be turned on / off for each macroblock The filter 117 is configured.

  When this algorithm is realized by dedicated hardware and software, the encoding control unit 111, the conversion unit 112, the quantization unit 113, the inverse quantization unit 114, the inverse conversion unit 115, and the motion compensation delay function, which are tools, are provided. Each of the memory 116 and the loop filter 117 possessed has dedicated hardware and software.

  FIG. 14 is a block diagram of the decoder 108. The decoder 108 is a part included in the encoding unit 118 of FIG. 13, and includes an inverse quantization unit 114, an inverse conversion unit 115, a memory 116 (having a variable delay function for motion compensation), and an in-loop filter. 117.

  The encoded data is inversely quantized by the inverse quantization unit 114, and inverse DCT transformed by the inverse transform unit 115 and decoded. The memory 116 and the in-loop filter 117 are used when decoding motion compensation prediction encoded data.

  JPEG, H.I. If several types of algorithms are to be processed by a method of encoding with a certain algorithm such as H.261, MPEG1, MPEG2, etc., hardware and software for realizing each algorithm are required.

  In one terminal, for example H.264. When a moving image is encoded by H.261 and a still image is encoded by JPEG, the configuration is as shown in FIG. That is, this encoder is H.264. 261 encoder 120 and JPEG encoder 121.

  Similarly, when the flexible encoding algorithm shown in FIG. There are several types of tools for the transform unit, quantization unit, inverse quantization unit, and inverse transform unit in H.261.

  Accordingly, the encoding unit 118 including the conversion unit 112, the quantization unit 113, the inverse quantization unit 114, and the inverse conversion unit 115 illustrated in FIG. 13 is configured by each tool illustrated in FIG. The decoding unit 119 including the inverse quantization unit 114 and the inverse transformation unit 115 illustrated in FIG. 14 is the tool configuration of the decoding unit 122 illustrated in FIG. 16.

  As the operation, tool information such as motion compensation tool A, inverse transform tool B, motion compensation tool C, inverse transform tool D, and quantization tool E shown in FIG. Certain motion vector information 101, conversion coefficient 102, motion vector information 103, and conversion coefficient 104 are transmitted to each tool.

  The control unit 123 performs selection control of which tool is used from each tool information, and each data is processed and decoded by the tool selected by the control unit 123.

  However, in this method, dedicated hardware and software must be prepared for each tool, which increases the scale of the decoder. In addition, when data processed by a tool not prepared by the decoder is received, it cannot be decoded. In order to solve this, it is conceivable that the received part is compiled to generate a processing program and decrypted by the general-purpose arithmetic processing unit.

  When the information shown in FIG. 11B is decoded by the general-purpose arithmetic processing unit and the compiler, the decoder shown in FIG. 17 is used. Tool information such as motion compensation tool A, inverse transform tool B, motion compensation tool C, transform tool D, and quantization tool E shown in FIG. 11B is transmitted to the compiler 125, and motion vectors that are subsequent data Information 101, transform coefficient 102, motion vector information 103, transform coefficient 104, and quantization step 105 are transmitted to general-purpose arithmetic processing unit 124.

In the compiler 125, a processing program for the general-purpose arithmetic processing unit 124 is generated, and the data transmitted next is processed by the general-purpose arithmetic processing unit 124 and decoded. Furthermore, the processing program once generated is stored and reused at the next decoding.
Japanese Patent Laid-Open No. 4-8064

  If the processing capability of the decoding side that processes an algorithm is lower than the sum of the processing capabilities of the tools that make up the algorithm requested by the encoding side, the tools sent from the encoding side are accumulated on the decoding side. However, since the processing capacity on the decoding side is low, the received data cannot be accurately decoded, and the memory of the tool storage unit is wasted.

  In addition, in the conventional encoding device and decoding device, when comparing the tool used on the encoding side with the tool stored on the decoding side, the tool itself must be compared. There was a problem that it took a very long time.

  When decoding encoded information using a new algorithm, even if the tools that make up the algorithm are the same as the tools that are already stored, the tool must be received again, and the transmission and reception time is considerably long. There was a problem.

  An object of the present invention is to provide a moving image encoding apparatus that can be decoded by a tool for decoding encoded data.

  Another object of the present invention is to provide a moving picture decoding apparatus capable of decoding with a tool for decoding encoded data from among a plurality of decoding tools.

According to an aspect of the present invention, motion-out compensation unit, the conversion unit, the inverse transform unit, a moving picture encoding apparatus including a control unit encoder comprising a quantization unit, the inverse quantization unit, motion The compensation unit operates using the first motion compensation tool or the second motion compensation tool , and key information indicating one of the first and second motion compensation tools used by the motion compensation unit is stored in the control unit. The control is transmitted to a decoding device that decodes the encoded data before the encoded data separately from the encoded data of the moving image, and the encoder encodes the moving image based on the key information. To do.

According to the present invention, the key information indicating one of the first and second motion compensation tools used by the motion compensation unit is transmitted to the decoding device that decodes the encoded data before the encoded data. Therefore, in the decoding apparatus, it is possible to recognize in advance which tool is used before receiving the encoded data. Therefore, before receiving the encoded data, the decoding apparatus can make the motion compensation tool used for decoding the encoded data ready. As a result, the decoding process can be performed efficiently .

Preferably, the first motion compensation tool and the second motion compensation tool have different pixel accuracy for motion compensation .

According to the present invention, the encoder can encode a moving picture with motion compensation suitable for the pixel accuracy of motion compensation, and can transmit key information indicating the pixel accuracy of motion compensation used for encoding. A decoder that decodes the encoded data can decode the encoded data with motion compensation adapted to the pixel accuracy of motion compensation.

According to another aspect of the present invention, there is provided a video decoding device including a decoder including a motion compensation unit, an inverse transform unit, and an inverse quantization unit, and a control unit , wherein the motion compensation unit is a first motion compensation tool. Alternatively, key information indicating one of the first and second motion compensation tools that operate using the second motion compensation tool and is used when the encoded data of the moving image is generated by the encoding device. The key information transmitted by the encoding device is received before the encoded data separately from the encoded data under the control of the control unit, and the decoder decodes the encoded data based on the key information. It is characterized by that.

According to the present invention, the key information indicating one of the first and second motion compensation tools used when the encoded data of the moving image is generated by the encoding device is separated from the encoded data. Received before encoded data. Therefore, it is possible to recognize in advance what kind of tool is used before receiving encoded data. Therefore, before receiving the encoded data, the motion compensation tool used for decoding the encoded data can be made ready. As a result, the decoding process can be performed efficiently .

Preferably, the first motion compensation tool and the second motion compensation tool have different pixel accuracy for motion compensation .

  According to the present invention, the decoder can perform decoding with motion compensation adapted to the pixel accuracy of motion compensation.

  Embodiments of the present invention will be described below with reference to the drawings.

(First embodiment)
FIG. 1 is a block diagram illustrating an encoding / decoding device according to a first embodiment of the present invention. The encoding / decoding device includes a control unit 11, an encoder 12, a response control unit 13, a decoder 14, a tool storage control unit 15, a tool storage unit 16, an encoding processing capability value storage unit 17, and a decoding processing capability value. The storage unit 18 and the ability value comparison unit 19 are provided.

  The control unit 11 is a part that controls the entire apparatus, and accumulates the sum of the processing capability values of the decoder 14 for each tool constituting the algorithm in the decoding processing capability value storage unit 18 as a decoding processing capability value in advance. Keep it. The processing capability value of the decoder 14 in units of tools is digitized. The encoder 12 adds the processing capability value necessary for the processing of the tool to each tool unit, and transmits it to the counterpart device.

  Here, quantification of the capability of the tool is performed by setting a standard process and a processing capacity required to perform the process, and comparing the processing load and the processing capacity of the process with other processes. .

  The encoder 12 is not digitized every time it is transmitted, but is set in advance by a tool creator or a selector based on a processing load or the like. The digitized capability is sent from the control unit 11 to the response control unit, and is multiplexed with encoded data and tool information and transmitted.

  The encoding processing capability value received by the decoder 14 is stored in the encoding processing capability value storage unit 17 for each tool constituting the algorithm. The sum of the output of the decoding processing capability value storage unit 18 and the processing capability value of the tool constituting the algorithm is input to the capability value comparison unit 19.

  The capability value comparison unit 19 compares each value and determines whether or not the processing of the algorithm using the received tool is possible in the own device. The determination can be made with a simple size comparison circuit. If it is determined from the comparison of the processing capability values that the processing is possible, a permission signal is sent to the tool storage control unit 15, the received tool is stored in the tool storage unit 16, and used thereafter for decoding the received data.

  If it is determined from the comparison of the processing capability values that the decoding processing capability is low and the processing is impossible even if the tools are accumulated, the received tool is unnecessary, so the accumulation impossible signal is sent to the tool accumulation control unit 15 to accumulate the tools. It does not accumulate in the part 16. At the same time, the response control unit 31 downloads the received tool and sends an instruction to send confirmation of whether or not processing is possible.

  Normally, under an encoding method in which the types of tools that make up the algorithm are specified, the mutual decoding capabilities are exchanged at the initial stage of communication to obtain the decoding capability of the other party, and then the data Encoding and transmission can begin.

  However, in the case of a method in which the encoding side and a tool that constitutes an algorithm that is a means for decoding the information are simultaneously transmitted from the encoding side, the load of the decoding process varies significantly depending on the type of tool to be selected. In such a capability exchange, it is considered that it is not easy to strictly obtain the decoding capability of the counterpart device in advance. Therefore, each time a new tool is received in this way, the ability values are compared, and a determination result as to whether or not processing is possible is transmitted to the partner apparatus.

  FIG. 2 is an explanatory diagram illustrating an operation example of the present embodiment. Operations of the encoding device 20 and the decoding device 21 will be described. The encoding device 20 and the decoding device 21 are the encoding / decoding devices shown in FIG. 1, and function as an encoding device and a decoding device, respectively.

  First, it is assumed that the encoding device 20 transmits the tool a and the encoding capability value Ca necessary for the processing of the tool a to the other party. The decoding device 21 accumulates the processing capability value Ca of the transmitted tool in the encoding processing capability value accumulation unit 17. Since the ability values of the tools constituting the algorithm are accumulated for each tool, if Cz is the sum of ability values other than Ca, Ca + Cz is necessary to process an algorithm that uses a tool having the ability value of Ca. Ability value.

  By comparing this value with the decoding capability value Cr set in advance in the decoding device 21, if Ca + Cz ≦ Cr, it can be determined that the newly received tool a can be used on the decoding side. If it is usable, the received tool is downloaded, and a download completion response is transmitted to the partner apparatus. As a result, the encoding device 20 can determine that data transmission using the transmitted tool is possible.

  Next, it is assumed that the encoding device 20 transmits the tool b and the encoding capability value Cb necessary for the processing of the tool b to the other party. The decoding device 21 accumulates the processing capability value Cb of the transmitted tool in the encoding processing capability value accumulation unit 17. Similar to the above-described operation example, Cb + Cz is a capability value necessary for processing an algorithm that uses a tool having a capability value of Cb.

  By comparing this value with a preset decoding capability value Cr, if Cb + Cz> Cr, it can be determined that the newly received tool cannot be used on the decoding side. In this case, the received tool is not downloaded and a download error response is transmitted to the partner apparatus. As a result, the encoding device 20 can determine that data transmission using the transmitted tool is impossible.

  Therefore, the encoding device 20 can decode data by retransmitting another new tool or transmitting data using the previously transmitted tool a.

  Thus, by making the tool correspondence information the processing capability of the tool and selecting a tool that falls within the decoding processing range, efficient decoding processing can be performed. In this case, the capability comparison can be performed quickly and efficiently by digitizing the processing capability value and transmitting it. Furthermore, since it is determined whether or not the receiving tool can be used based on the processing capability, and the tool is downloaded only when it is used, the processing time required for transmission and reception can be shortened.

(Second Embodiment)
FIG. 3 shows an example of tool management in the second embodiment of the present invention. As shown in FIG. 3, a unique key is attached to each tool. For example, keys corresponding to all tools are set so that the key is Aa01 and the corresponding tool is integer pixel precision motion compensation, so that each tool can be identified.

  FIG. 4 shows an example of this key setting. For example, as shown in FIG. 4, the key represents a large classification (motion compensation prediction, DCT conversion, etc.) in uppercase letters, a small classification (integer pixel precision, half-pixel precision, etc.) in lowercase letters, and a version in two numbers. Is done.

  FIG. 5 is a block diagram of the encoding apparatus according to this embodiment. The encoding device stores a control unit 31 that controls the entire device, a tool storage unit 33 that stores each tool, and a tool processing unit 33 that stores a tool processing program corresponding to each key. A key table 32, a processing operation unit 34 that reads and executes a tool processing program from the tool storage unit, a work memory 35 used in the processing operation unit 34, and a network interface 36 that communicates with the outside. Yes.

  FIG. 6 is a block diagram of the decoding apparatus according to this embodiment. This decryption apparatus includes a control unit 41 that controls the entire apparatus, a key buffer 47 that temporarily stores received keys, a tool storage unit 43 that stores tools, and a processing program for tools corresponding to each key. A key table 42 for storing where the data is stored in the storage unit 43; a processing calculation unit 44 that reads and executes a tool processing program from the tool storage unit 43; a work memory 45 used by the processing calculation unit 44; A network interface 46 for performing communication with the outside is provided.

  FIG. 7 shows an example of the storage state of each tool in the tool storage unit 33 of the encoding device of FIG. 5 and the tool storage unit 43 of the decoding device of FIG. FIG. 8 shows an example of correspondence between each key in the key table 32 in FIG. 5 and the key table 42 in FIG. 6 and the storage location in the tool storage units 33 and 43 of the tool corresponding to the key. .

  The encoding device in FIG. 5 and the decoding device in FIG. 6 are connected to various networks such as ISDN and wireless network by network interfaces 36 and 46, respectively, as shown in FIG.

  FIG. 10 is a flowchart showing an example of a communication procedure when the encoding device of FIG. 5 and the decoding device of FIG. 6 are connected as shown in FIG. First, an algorithm used in the encoding device and a tool constituting the algorithm are determined (step S1). Then, the key of each tool is transmitted to the decryption device (step S2).

  The decryption device receives the key and takes the key into the key buffer 47 (step T1). Then, the keys fetched into the key buffer 47 are checked against the key table 42 (step T2), and if not registered, these keys are transmitted to the encoding device together with the tool transfer request (step T3).

  The encoding device transmits a tool processing program corresponding to the key to the decoding device (step S4). The decryption apparatus stores the processing program for the tool in the tool storage unit 43, and registers the storage address and key in the key table 42 (step T5). Then, the tool is transferred from the tool storage unit 43 to the calculation unit 44 while referring to the key table 42 (step T6).

  Thereafter, the decoding apparatus transmits an encoded data transfer request to the encoding apparatus (step T7), and the encoding apparatus that has received the request transmits the encoded data to the decoding apparatus (steps S5 and S6).

  The decoding device receives the encoded data (step T8) and decodes it (step T9). When the transmission of all the encoded data is completed, the encoding device transmits an encoded data transmission end signal to the decoding device (steps S7 and S8). When the decoding device receives the signal (step T10), the communication is performed. Ends.

  In this way, by comparing the keys using the keys unique to the tools, the corresponding tools can be easily compared and selected, and the processing time can be shortened. Furthermore, it is possible to reduce the processing time required for transmission and reception by determining whether or not the same tool is present in the decryption apparatus using the key and transferring the tool only when there is not.

  The decoding apparatus according to the present embodiment configures a decoding algorithm only from tools necessary for decoding received encoded data by comparing and selecting an optimal tool based on information indicating the received tool. Therefore, decoding processing is efficient and processing can be performed in a short time.

  Further, by comparing information indicating the tools, it is not necessary to compare the tools themselves, and quick tool selection processing is possible. In particular, by using the information indicating the tool as a key unique to the tool, the process of comparing and selecting the tool in the tool storage means can be efficiently performed in a short time.

It is a block diagram which shows the encoding / decoding apparatus which concerns on 1st Embodiment of this invention. It is explanatory drawing which shows operation | movement of the encoding / decoding apparatus which concerns on 1st Embodiment of this invention. It is explanatory drawing which shows an example of the tool corresponding to the key in 2nd Embodiment of this invention. It is explanatory drawing which shows an example of the key in 2nd Embodiment of this invention. It is a block diagram which shows the encoding apparatus which concerns on 2nd Embodiment of this invention. It is a block diagram which shows the decoding apparatus which concerns on 2nd Embodiment of this invention. It is explanatory drawing which shows an example of the storage state of each tool in 2nd Embodiment of this invention. It is explanatory drawing which shows the storage state of the key table in 2nd Embodiment of this invention. It is explanatory drawing of the network connection in 2nd Embodiment of this invention. It is a flowchart which shows the communication procedure of the encoding apparatus and decoding apparatus in 2nd Embodiment of this invention. It is explanatory drawing which shows the data sequence of the encoding output information in the conventional encoding / decoding apparatus. It is a block diagram which shows the conventional encoding / decoding apparatus. It is a block diagram which shows the encoder in the encoding / decoding apparatus of FIG. It is a block diagram which shows the decoder in the encoding / decoding apparatus of FIG. H. 2 is a block diagram illustrating an encoder that performs encoding with H.261 and JPEG. FIG. It is a block diagram which shows each tool of a decoder. It is a block diagram which shows the decoder which performs the process by a compiler.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 11 Control part, 12 Encoder, 13 Response control part, 14 Decoder, 15 Tool accumulation | storage control part, 16 Tool accumulation | storage part, 17 Encoding processing capability value accumulation | storage part, 18 Decoding processing capability value accumulation | storage part, 19 Competence value comparison Department.

Claims (4)

A video encoding device including an encoder and a control unit including a motion compensation unit, a conversion unit, an inverse conversion unit, a quantization unit, and an inverse quantization unit,
The motion compensation unit operates using a first motion compensation tool or a second motion compensation tool ;
The key information indicating one of the first and second motion compensation tools used by the motion compensation unit is controlled by the control unit, before the encoded data, separately from the encoded data of the moving image. Transmitted to a decoding device for decoding the encoded data ,
The moving image encoding apparatus, wherein the encoder encodes a moving image based on the key information . The moving image coding apparatus according to claim 1, wherein the first motion compensation tool and the second motion compensation tool have different pixel accuracy for motion compensation . A video decoding device including a motion compensation unit, an inverse transform unit, a decoder including an inverse quantization unit and a control unit ,
The motion compensation unit operates using a first motion compensation tool or a second motion compensation tool ;
Key information indicating one of the first and second motion compensation tools used when the encoded data of the moving image is generated by the encoding device, the key information transmitted by the encoding device , under the control of the control unit, and the encoded data separately, it is received before the encoded data,
The moving picture decoding apparatus, wherein the decoder decodes the encoded data based on the key information . The moving image decoding apparatus according to claim 3, wherein the first motion compensation tool and the second motion compensation tool have different pixel accuracy for motion compensation .

Images