JP2005277486A - Video coding arithmetic amount control method, video coding arithmetic amount control apparatus, video coding arithmetic amount control program, and recording medium for recording the program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an arithmetic amount control capable of preventing occurrence of a defect of the code amount control by the arithmetic amount control and deterioration in the image quality by taking into account characteristics of an arithmetic amount variable element in the case of controlling the arithmetic amount in video coding processing. <P>SOLUTION: The arithmetic amount variable element in the video coding processing is classified into two types: a generated code amount increasing type and a generated code amount decreasing type. A flowchart for explaining the arithmetic amount control includes: a step S2 of acquiring code amount control information when the reduction in the arithmetic amount is required (S1); a step S3 of checking which of the increase and decrease in the generated code amount the code amount control algorithm controls; a step S4 of utilizing the arithmetic amount variable element of the generated code amount increase time in the case of the algorithm in the increase tendency to carry out the arithmetic amount control; a step S7 of carrying out the arithmetic amount control by utilizing the arithmetic amount variable element of the generated code decrease time in the case of the algorithm in the decrease tendency to carry out the arithmetic amount control; and steps S6, S9 of utilizing also arithmetic amount variable elements of different types to carry out the arithmetic amount control when the arithmetic amount reduction amount is insufficient only with the respective arithmetic amount variable element types (S5, S8). <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a technique for controlling the amount of calculation of encoding processing according to processing time, system load, and the like in video encoding processing.

  Currently, an encoding method is often used in which a video signal is converted into a digital signal and encoded to compress the data size. In particular, in recent years, encoding processing by software using a personal computer or the like is frequently performed. However, in order to obtain sufficient encoding efficiency, with the current CPU power, the calculation processing time of the encoding process requires a time that is several tens of times longer than the encoding target video time. Therefore, in order to perform the encoding process in a somewhat realistic time, it is necessary to control the calculation amount. This is especially important for applications that require real-time processing such as live broadcasts and videophones.

  In the calculation amount control of the encoding process, processing restriction, truncation, processing algorithm switching, and the like are mainly performed on a calculation amount control target (hereinafter referred to as a calculation amount variable element). For example, in the motion vector search process, the amount of calculation can be controlled by enlarging / reducing the search range or changing the detection accuracy. In addition, in DCT and quantization processing, the amount of calculation can be controlled by using a method that terminates processing of higher-order coefficients. When it is necessary to reduce the amount of calculation, the amount of calculation is controlled using such a variable amount of calculation.

  As a method of controlling the amount of encoding calculation, [1] a control method using a CPU load (for example, refer to Patent Document 1 “Image encoding method by software”), [2] a video signal is stored in a buffer memory, There is a method (for example, refer to “Patent Document 2“ Image Encoding and Decoding Method ”) for controlling the amount of calculation with the occupied amount. [3] A method for determining the amount of calculation so that the processing can be completed within the target time from the past processing time (see, for example, Patent Document 3 “Processing Time Adaptive Image Coding Method and Program Recording Medium”). is there.

  [1] In the calculation amount control method according to the CPU load, when the CPU load increases, control is performed so as to reduce the calculation amount for the calculation amount variable element. Conversely, when the load on the CPU is light, control is performed to increase the amount of computation.

  FIG. 5 is a diagram illustrating a configuration example of a conventional video encoding device. The video encoding apparatus includes a calculation amount control unit 50, a frame memory (FM) 51, a motion detection unit 52, a motion compensation unit 53, a subtractor 54, a DCT unit 55, a quantization unit 56, an inverse quantization unit 57, and an IDCT unit. 58, an adder 59, a clipping unit 60, an encoding unit 61, and a code amount control unit 62.

  The motion detection unit 52 searches for a motion vector between the image (encoded block) of the input video signal and the reference image stored in the frame memory 51, and the motion compensation unit 53 performs motion compensation using the detected motion vector. To create a predicted image from the reference image. The subtractor 54 calculates a prediction error from the difference between the image of the input video signal and the prediction image, and the DCT unit 55 performs a discrete cosine transform on the prediction error. The quantization unit 56 quantizes the transform coefficient of the discrete cosine transform according to the quantization step size determined by the code amount control unit 62, and the encoding unit 61 encodes the quantized transform coefficient. The encoding unit 61 encodes the motion vector detected by the motion detection unit 52.

  The inverse quantization unit 57 inversely quantizes the transform coefficient quantized by the quantization unit 56, and the IDCT unit 58 performs inverse DCT transform on the inversely quantized transform coefficient. The adder 59 adds the predicted image and the prediction error resulting from the inverse DCT transformation to generate a local decoded image, and the clipping unit 60 performs clipping so that the value of the decoded image falls within a predetermined range. The image is stored in the frame memory 51 as an image. The code amount control unit 62 determines the quantization step size from the target code amount and the actual code amount encoded by the encoding unit 61, and controls the code amount by changing the quantization step size. The calculation amount control unit 50 controls the calculation amount for the calculation amount variable element based on the CPU load information and the like.

  In the example of FIG. 5, for the purpose of encoding processing in real time, ON / OFF of motion search and arithmetic termination of higher-order coefficients of DCT coefficients are used as calculation amount variable elements. The calculation amount control unit 50 realizes calculation amount control by switching the operation of the two calculation amount variable elements according to the CPU load. When there is a margin in the CPU load, both calculation amount variable elements perform the original operation, that is, the operation that maximizes the coding efficiency without limiting the calculation amount. However, when the CPU load is high and real-time processing is difficult, the operation of the calculation amount variable element is switched to reduce the calculation amount according to the load level.

  At this time, which operation amount variable element to switch the operation of is determined from the CPU load height and the amount of operation amount reduction of each operation amount variable element. For example, if real-time processing is possible only by the DCT coefficient truncation process, the DCT coefficient truncation process is performed without changing the motion search process. When the CPU load is very high, the motion search is turned off and the DCT coefficient truncation process is performed simultaneously.

  [2] In the calculation amount control method corresponding to the input buffer occupation amount, an input buffer for accumulating uncoded video signals is provided, and the calculation amount control is performed based on the input buffer occupation amount. When the occupation amount of the input buffer increases and the processing speed is judged to be slow with respect to the input information amount, the calculation amount is reduced. Conversely, when the input buffer occupancy decreases and the processing speed is high relative to the information amount of the input signal, control is performed to increase the amount of computation.

  [3] In the calculation amount control method based on the processing time prediction, the calculation amount of the encoding process required for the next encoding target signal is estimated from the past encoding processing time, and the processing content is determined.

Conventionally, the calculation amount of the encoding process has been controlled by the method described above.
Japanese Patent No. 3283159 JP-A-9-46708 JP 2002-112274 A

  In many coding apparatuses, a target generated code amount is given and code amount control is performed. The calculation amount control described above operates independently of the code amount control.

  The variable amount of calculation used for calculation control is classified into one of the following types.

• Generated code amount increase type Generated code amount increases by reducing the amount of computation. For example, if the search range of motion vector search is narrowed or the detection accuracy is reduced, the amount of code is increased due to a decrease in the efficiency of motion compensated prediction although the amount of computation is reduced.

• Generated code amount reduction type Generated code amount decreases by reducing the amount of computation. For example, the generated code amount is reduced by omitting the calculation of higher-order coefficients in the DCT unit 55 and the quantization unit 56 or forcibly stopping the encoding process like macro block skipping. However, since the information of the encoding target signal is not transmitted, the image quality is degraded.

  FIG. 6 is a diagram illustrating the relationship between the encoding efficiency and the amount of calculation. FIG. 6A shows the relationship between the calculation amount and the generated code amount, and FIG. 6B shows the relationship between the calculation amount and the coding distortion amount. With either type of variable calculation amount, encoding efficiency decreases when the calculation amount is reduced. However, the effect of the decrease in coding efficiency on the amount of generated code is different.

  In the generated code amount increasing type, although the coding distortion amount does not change greatly, the generated code amount increases. In the generated code amount reduction type, although the generated code amount is reduced, necessary information is not encoded, so that the encoding distortion amount is increased.

  In the conventional calculation amount control method, the calculation amount control unit 50 and the code amount control unit 62 operate independently, and do not take into account the characteristics of the above-described calculation amount variable element, but only control the target calculation amount. Was aimed at. For this reason, in the generated code amount control, there is a possibility that it may become uncontrollable. For example, when the code amount control unit 62 works in a direction to decrease the generated code amount and the calculation amount control unit 50 works in a direction to increase the generated code amount, the generated code amount is predicted as the code amount control unit 62 predicts. Will not decrease. If this state continues, the actual generated code amount may exceed the target generated code amount.

  In addition, when the actual generated code amount is small relative to the target generated code amount and there is no allowance for the amount of calculation, using the generated code amount decreasing type element as the amount control element will affect the effect of the amount control. Causes significant image quality degradation.

  As described above, in the conventional method, the control of the generated code amount may not function because the characteristics of the variable amount of calculation are ignored and the two control algorithms operate independently. In addition, degradation of image quality due to calculation amount control also appears.

  An object of the present invention is to solve the above-described problems and to realize effective calculation amount control capable of preventing failure of code amount control and deterioration of image quality.

  In order to solve the above-described problem, the present invention is characterized in that, when performing computation amount control, the type of computation amount variable element used for computation amount control is switched in consideration of the result of code amount control. If the result of code amount control is to reduce the generated code amount, the amount of generated code amount reduction type variable is used as the variable for the amount of calculation, and if the generated code amount is increased, the amount of calculation As the variable element, a generated code amount increasing type calculation amount variable element is used.

  That is, the present invention controls the amount of computation according to processing time, system load, etc. by controlling a plurality of computation amount variable elements that can reduce the amount of computation by changing the processing content in the encoding process. In the video coding system, the code amount control includes means for acquiring control information for code amount control, and means for switching a variable amount of calculation used for reducing the amount of calculation based on the control information for code amount control. The calculation amount control is performed in cooperation with the above.

  Further, according to the present invention, when the amount of calculation variable element is classified into a generated code amount increasing type and a generated code amount decreasing type, and the code amount control is controlled to increase, the generated code amount increasing type of calculation amount variable When the code amount control is controlled so that the code amount control tends to decrease, the generated code amount decrease type operation amount variable element is preferentially used to reduce the operation amount.

  In addition, the present invention compares the target calculation amount reduction amount with the expected calculation amount reduction amount after determining the type of the calculation amount variable element to be used from the code amount control information, and determines the expected calculation amount reduction amount as the target. In the case where the amount of calculation reduction is not satisfied, the amount of calculation is reduced using different types of calculation amount variable elements.

  As the control information of the code amount control, for example, the quantization step size or the change amount of the buffer occupation amount, the difference between the target generated code amount and the current generated code amount can be used.

  The processing for controlling the amount of computation in the above video encoding can be realized by a computer and a software program, and the program can be recorded on a computer-readable recording medium or provided through a network. .

  FIG. 1 is a flowchart illustrating calculation amount control according to the present invention.

[Step S1: Determination of necessity of calculation amount control]
The necessity of the calculation amount control is confirmed, and if necessary, the process proceeds to the next step S2. If unnecessary, the following processing is not performed.

[Step S2: Acquisition of Code Quantity Control Information]
Code amount control information is acquired.

[Step S3: Confirmation of Trend of Generated Code Amount]
Next, it is confirmed from the obtained code amount control information whether the code amount control algorithm is controlled to increase or decrease. If there is an increasing trend, the process proceeds to step S4. If it is decreasing, the process proceeds to step S7.

[Step S4: Use Generated Code Quantity Increasing Calculation Variable Variable]
In the case of an increasing tendency, the calculation amount variable element type to be used is determined to be the generated code amount increasing type. That is, the type of calculation variable is switched to the generated code amount increasing type, and the calculation amount control is attempted using only the generated code amount increasing type of calculation variable variable.

[Step S5: Confirmation of Calculation Amount Reduction]
It is determined whether the amount of calculation amount reduction is sufficient with only the generated code amount increase type calculation amount variable element used. If not enough, the process proceeds to step S6.

[Step S6: Use the generated code amount reduction type calculation amount variable element]
When the amount of calculation amount reduction is insufficient, the amount of calculation variable element type to be used is added, and the amount of calculation control is performed using the generation amount variable element of the generated code amount reduction type.

[Step S7: Utilizing Generated Code Amount Decreasing Type Variable of Calculation Amount]
As a result of confirming the tendency of the generated code amount in step S3, if the tendency is a decrease, the calculation amount variable element type to be used is determined to be the generated code amount decreasing type. In other words, the variable of the amount of calculation variable element is switched to the generated code amount reduction type, and the calculation amount control is attempted using only the generation amount variable element of the generated code amount reduction type.

[Step S8: Confirmation of Calculation Amount Reduction]
It is determined whether the amount of calculation amount reduction is sufficient with only the generated code amount reduction type calculation amount variable element used. If not enough, the process proceeds to step S9.

[Step S9: Use the generated code amount increasing type calculation amount variable element]
When the amount of calculation reduction is insufficient, a calculation amount variable element type to be used is added, and the calculation amount control is performed using the generated code amount increase type calculation amount variable element.

  As described above, basically, the type of calculation amount variable element is switched according to the code amount control information. However, when the amount of calculation amount reduction is insufficient, a different type of calculation amount variable element is also used. Here, as the code amount control information, a quantization step size, a change amount of the buffer occupation amount, a difference between the target generated code amount and the actual generated code amount, and the like can be used.

  By switching the type of the calculation amount variable element according to the result of the code amount control, the calculation amount control can be performed without affecting the code amount control.

  According to the present invention, it is possible to prevent the code amount control from failing by performing the operation amount control by switching the type of the operation amount variable element from the result of the code amount control. Further, it is possible to prevent the image quality from being deteriorated by the calculation amount control when the generated code amount is small.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 2 is a diagram illustrating a configuration example of the video encoding device according to the present embodiment. The video encoding apparatus includes a calculation amount control unit 10, a frame memory (FM) 11, a motion detection unit 12, a motion compensation unit 13, a subtractor 14, a DCT unit 15, a quantization unit 16, an inverse quantization unit 17, and an IDCT unit. 18, an adder 19, a clipping unit 20, an encoding unit 21, and a code amount control unit 22.

  In this embodiment, a motion compensated interframe predictive coding method and a discrete cosine transform (DCT) are used as coding algorithms. The motion detection unit 12 searches for a motion vector between the image of the input video signal and the reference image stored in the frame memory 11, and the motion compensation unit 13 performs motion compensation using the detected motion vector, A predicted image is created from the reference image. The subtractor 14 calculates a prediction error from the difference between the image of the input video signal and the predicted image, and the DCT unit 15 performs a discrete cosine transform on the prediction error. The quantization unit 16 quantizes the transform coefficient of the discrete cosine transform according to the quantization step size determined by the code amount control unit 22, and the encoding unit 21 encodes the quantized transform coefficient. The encoding unit 21 encodes the motion vector detected by the motion detection unit 12.

  The inverse quantization unit 17 inversely quantizes the transform coefficient quantized by the quantization unit 16, and the IDCT unit 18 performs inverse DCT transform on the inversely quantized transform coefficient. The adder 19 adds the prediction image and the prediction error resulting from the inverse DCT transformation to generate a local decoded image, and the clipping unit 20 performs clipping so that the value of the decoded image falls within a predetermined range. The image is stored in the frame memory 11 as an image. The code amount control unit 22 determines a quantization step size from the target code amount and the actual code amount encoded by the encoding unit 21, and controls the code amount by changing the quantization step size. The calculation amount control unit 10 controls the calculation amount for the calculation amount variable element according to the target calculation amount reduction amount calculated from the target processing time or the like based on the control information of the code amount control unit 22.

There are two types of calculation amount variable elements used in the present embodiment. As the generated code amount increase type calculation amount variable element, motion search processing in the motion detector 12 is used, and as the generated code amount decrease type calculation amount variable element, A DCT coefficient truncation process in the DCT unit 15 or the like was used. As the code amount control information, the quantization step size is used, and the increase / decrease tendency is determined by comparing with the immediately preceding quantization step size. Calculation amount control is performed by the difference of the actual processing time with respect to the target processing time T _o T _t.

Target processing time T _o is the target processing time for one frame T, one frame of the coded blocks N, when the encoded blocks to the current and M, for example, is a value obtained by the following equation .

_{T o = (M / N)} × T
The processing time T _t is the current time taken to encode data for M blocks from the start of encoding the current frame. Of course, the target processing time T _o and the actual processing time T _t may be calculated on the basis of, for example, the encoding start time of the input video instead of the frame unit.

  FIG. 3 is a diagram illustrating a configuration example of the calculation amount control unit in the present embodiment. The calculation amount control unit 10 includes a reduction calculation amount calculation unit 101, a generated code amount increase / decrease determination unit 102, a quantization step size memory 103, and a calculation amount variable element control unit 104.

The calculation amount control unit 10 is supplied with the current processing time T _t and the quantization step size QP _t as inputs. The output from the calculation amount control unit 10 is ON / OFF information of calculation amount reduction processing for each calculation amount variable element.

The reduction calculation amount calculation unit 101 calculates the calculation amount to be reduced from the current processing time T _t . This reduction calculation amount is sent to the calculation amount variable element control unit 104. The generated code amount increase / decrease determination unit 102 compares the input quantization step size QP _t with the immediately preceding quantization step size QP _t−1 and determines the control tendency of the code amount control unit 22 from the magnitude relationship. If the quantization step size is small, it is determined that the generated code amount is increasing, and conversely if it is large, it is determined that the code is decreasing. The input quantization step size QP _t is stored in the quantization step size memory 103 for the next determination.

  The calculation amount variable element control unit 104 controls whether or not to perform motion search processing and DCT coefficient truncation processing. By turning on the output of the calculation amount variable element control signal, the calculation amount reduction processing is performed. That is, when the output from the calculation amount variable element control unit 104 to the motion detection unit 12 is turned ON, the motion detection unit 12 skips without performing motion detection (motion search OFF), and the DCT unit 15 / IDCT unit 18 / When the output to the quantization unit 16 / inverse quantization unit 17 is turned ON, the DCT unit 15 / IDCT unit 18 / quantization unit 16 / inverse quantization unit 17 does not calculate higher-order coefficients (DCT coefficient truncation) processing). When it is not necessary to reduce the calculation amount, the output to both calculation amount variable elements is turned off regardless of the result of the generated code amount increase / decrease determination unit 102. When it is necessary to reduce the amount of calculation, based on the determination result of the generated code amount increase / decrease determination unit 102, the control signal of one of the calculation amount variable elements is turned ON. In addition, when only one of the calculation amount variable elements is less than the target calculation amount reduction amount, the control signals of both calculation amount variable elements are turned ON.

FIG. 4 is a calculation amount control process flowchart in the present embodiment. Here, the calculation amount reduction amount by the motion search process is _TME, and the calculation amount reduction amount by the DCT coefficient truncation processing is _TDCT . These calculation amount reduction amounts T _ME and T _DCT can be obtained in advance by experiments using standard video signals.

First, the calculation amount control unit 10 acquires the current processing time T _t by the reduction calculation amount calculation unit 101 (step S10), and calculates the necessary reduction calculation time T _r (T _r = T _t −T _o ). (Step S11).

Next, the calculation amount control unit 10 determines whether or not calculation amount control is necessary depending on whether or not T _r > 0 (step S12). If T _r > 0 is not satisfied, the amount of calculation control is unnecessary, so the processing is terminated without reducing the amount of calculation, and the normal encoding process is continued.

If T _r > 0 in step S12, the generated code amount increase / decrease determination unit 102 acquires the current quantization step size QP _t (step S13), and the current quantization step size QP _t and the quantization step The past quantization step size QP _t−1 stored in the size memory 103 is compared (step S14), and the increasing / decreasing tendency of the generated code amount is confirmed. In accordance with this increase / decrease trend, the calculation amount variable element control unit 104 switches the calculation amount variable element as described below.

If QP _t ≦ QP _t−1 in step S14, motion search is skipped (motion search OFF) (step S15). The reduction calculation time T _r is compared with the calculation amount reduction amount T _ME by the motion search process (step S16), and if T _r > T _ME , further DCT coefficient truncation processing is performed (step S17). If T _r ≦ _TME , it is determined that the amount of calculation amount reduction is sufficient, and the process is terminated.

If QP _t ≦ QP _t−1 is not satisfied in step S14, DCT coefficient truncation processing is performed (step S18). The reduction calculation time T _r and the calculation amount reduction amount T _DCT by the DCT coefficient truncation process are compared (step S19). If T _r > T _DCT , the motion search is further skipped (motion search OFF) (step S19). S20). If T _r > T _DCT is not satisfied, it is determined that the amount of calculation reduction is sufficient, and the process is terminated.

In this way, by extracting the rate control information from the quantization step size QP _t switch the type of operation the variable element, it is possible to Computational Complexity Reduction in line with code amount control. Although an example in which the quantization step size is used as the code amount control information has been described, information such as a change amount of the buffer occupancy amount, a difference between the target generated code amount and the current generated code amount may be used.

  In the present embodiment, there are two calculation amount variable elements, but more calculation amount variable elements can be controlled by using more calculation amount variable elements. For example, instead of performing calculation amount control on whether or not to skip the motion search by the motion detection unit 12, to what extent the motion search range of the motion detection unit 12 is reduced, and to what extent the detection accuracy is reduced It is also possible to perform stepwise calculation amount control.

It is a flowchart explaining the calculation amount control by this invention. It is a figure which shows the structural example of the video coding apparatus in this Embodiment. It is a figure which shows the structural example of the calculation amount control part in this Embodiment. It is a calculation amount control processing flowchart in the present embodiment. It is a figure which shows the structural example of the conventional video coding apparatus. It is a figure which shows the relationship between encoding efficiency and the amount of calculations.

Explanation of symbols

10, 50 Computation amount controller 11, 51 Frame memory (FM)
12, 52 Motion detection unit 13, 53 Motion compensation unit 14, 54 Subtractor 15, 55 DCT unit 16, 56 Quantization unit 17, 57 Inverse quantization unit 18, 58 IDCT unit 19, 59 Adder 20, 60 Clipping unit 21, 61 Encoding unit 22, 62 Code amount control unit 101 Reduction calculation amount calculation unit 102 Generated code amount increase / decrease determination unit 103 Quantization step size memory 104 Calculation amount variable element control unit

Claims (10)

An amount-of-computation control method for controlling the amount of computation in encoding processing by controlling a plurality of computation amount variable elements capable of changing the amount of computation by changing processing contents in video encoding processing,
A process of acquiring control information of code amount control;
A process of determining a variable element for calculation amount to be used for control of calculation amount based on the control information of the code amount control;
A video encoding calculation amount control method comprising: controlling a calculation amount of an encoding process by controlling a determined calculation amount variable element. The video encoding calculation amount control method according to claim 1,
The plurality of calculation amount variable elements are classified into either a generated code amount increasing type in which the generated code amount increases when the calculation amount is reduced, or a generated code amount decreasing type in which the generated code amount decreases when the calculation amount is reduced. Has been
In the process of determining the calculation amount variable element used for the calculation amount control, it is determined from the control information of the code amount control whether the code amount is controlled to increase or decrease in the code amount control. If it is determined and controlled to increase, the generated code amount increasing type calculation amount variable element is preferentially used for control of the calculation amount and is controlled to decrease. In such a case, the video coding calculation amount control method is characterized in that the generated code amount reduction type calculation amount variable element is used as a calculation amount variable element that is preferentially used for controlling the calculation amount. The video encoding calculation amount control method according to claim 2,
In the process of determining the variable amount of calculation to be used for controlling the calculation amount, it is necessary to reduce the amount of calculation when the calculation amount is controlled for the variable amount of calculation that has been determined to be used preferentially. If the amount of calculation reduction is less than the target amount of calculation reduction, the generated code amount increase type calculation amount variable element determined to be used preferentially or A video encoding calculation amount control method, characterized in that a calculation amount variable element of a type different from one of the generated code amount reduction type calculation amount variable elements is also used as a calculation amount variable element used to control the calculation amount. In the video coding calculation amount control method according to any one of claims 1 to 3,
In the process of acquiring the control information of the code amount control, the change amount of the quantization step size or the buffer occupancy amount, or the information of the difference between the target generated code amount and the actual generated code amount is used as the control information of the encoding control. A video encoding calculation amount control method characterized by comprising: A video encoding calculation amount control device that controls the calculation amount of an encoding process by controlling a plurality of calculation amount variable elements that can change the calculation amount by changing the processing content in the video encoding process,
Means for acquiring control information of code amount control;
A calculation amount for controlling the calculation amount of the encoding process by determining a calculation amount variable element to be used for control of the calculation amount based on the control information of the code amount control and controlling the determined calculation amount variable element A video encoding calculation amount control device comprising: variable element control means. In the video coding calculation amount control device according to claim 5,
The plurality of calculation amount variable elements are classified into either a generated code amount increasing type in which the generated code amount increases when the calculation amount is reduced, or a generated code amount decreasing type in which the generated code amount decreases when the calculation amount is reduced. Has been
The calculation amount variable element control means determines from the control information of the code amount control whether the code amount is controlled to increase or decrease in code amount control, and is controlled to increase. If the generated code amount increasing type variable of the calculation amount is preferentially used for the control of the calculation amount, and if it is controlled to decrease, the generated code amount decreasing type of the calculation amount variable A video encoding calculation amount control device characterized in that an element is preferentially used for controlling the calculation amount. In the video coding calculation amount control device according to claim 6,
The calculation amount variable element control means compares the calculation amount reduction amount when the calculation amount control is performed on the calculation amount variable element determined to be used preferentially and the target calculation amount reduction amount that needs to be reduced. However, when the calculation amount reduction amount is less than the target calculation amount reduction amount, the generated code amount increase type calculation amount variable element or the generated code amount decrease type calculation amount that is determined to be preferentially used. A video encoding calculation amount control device characterized in that a calculation amount variable element of a type different from one of the variable elements is also used for control of the calculation amount. In the video coding calculation amount control device according to any one of claims 5 to 7,
The means for acquiring the control information of the code amount control includes, as encoding control information, information on a change in the quantization step size or the buffer occupancy amount, or information on a difference between the target generated code amount and the actual generated code amount. A video encoding calculation amount control device characterized by comprising:   In order to realize the video encoding calculation amount control device according to any one of claims 5 to 7, at least a means for acquiring control information of the code amount control and a calculation amount variable element control means as a computer Video encoding calculation amount control program for functioning. In order to realize the video encoding calculation amount control device according to any one of claims 5 to 7, at least a means for acquiring control information of the code amount control and a calculation amount variable element control means as a computer A video encoding calculation amount control program recording medium characterized in that a video encoding calculation amount control program for functioning is recorded on a computer-readable recording medium.

Images