JP2004173117A - Device and method for decoding encoded data - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device for decoding encoded data which can seamlessly reproduce an editing point of compressed encoded data. <P>SOLUTION: This decoding device reproduces data edited by connecting two pieces of compressed image encoded data with an optional frame. A decoding time allocating means 12 controls the time of a difference between the display time and the decoding time of image encoded data before the editing point, to be reproduced temporally ahead, so as to be allocated to the decoding of image encoded data after the editing point, to be reproduced temporally behind. In addition, a decoding position managing means 13 stores to which position of a frame the image encoded data after the editing point, to reproduced temporally behind, has been decoded by the preceding decoding and controls decoding so as to be started from the point of time when the preceding decoding has been completed in the next decoding. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a decoding device and a decoding method for decoding digitally encoded image encoded data, and more particularly to a reproduction control at an editing point in image encoded data.
[0002]
[Prior art]
Conventionally, as a coding method for compressing a digital image signal, an MPEG (Moving Picture Experts Group) method specified in ISO / IEC11172, 13818 has been proposed. FIG. 7 shows a typical data structure of the MPEG system. In this figure, a GOP (GroupOf Pictures), which is a unit of coding obtained by dividing one sequence of a moving image, is composed of 1 frame for an I picture, 4 frames for a P picture, and 10 frames for a B picture, for a total of 15 frames. . Here, the I picture is an intra-frame coded image, the P picture is an inter-frame forward prediction coded image, and the B picture is an inter-frame bidirectional predicted coded image. Specifically, as shown by the arrow in FIG. 7A, I2 is intra-frame coded only with the I2 frame, P5 is I2, and P8 is inter-frame coded with reference to P5. B0 and B1 are inter-frame predictive coded with reference to P'14 and I2, and B3 and B4 with reference to I2 and P5. Actual coded data is coded in the order shown in FIG. This is to make it possible to decode frames required for decoding first.
[0003]
Here, reproduction in the DVD video recording standard using the MPEG compression method will be described with reference to FIGS. In the DVD video recording standard, there is an entry point reproduction in which moving image data encoded using the MPEG compression technique is edited at an arbitrary point and images are connected, and FIG. Indicates frame skip. VOBU in the figure is an abbreviation of Video OBJECT Unit, and is a unit of a sequence composed of one or a plurality of GOPs. As shown in FIG. 8, when the reproduction is completed at P'5 of VOBU (Stop Point in the figure) and then edited to reproduce B6 of VOBU1 (Start Point in the figure), After the reproduction of P'5 of VOBU0 is completed, it is necessary to decode I2, P5 and P8 of VOBU1 in order to display B6. The display of the pictures included in the range of Skip in FIG. 8 is not performed. FIG. 9 is a timing chart of decoding in a general decoding device at an editing point. When the display of P'5 of VOBU0 is completed, the reproduction of the next VOBU is performed, and the display is switched to B6 when the decoding of I2, P5, and P8 is performed. In FIG. 9, SYNC is a vertical synchronization signal of video, and one unit is two vertical synchronization signals (2 V).
[0004]
[Patent Document 1]
DVD Specifications for Read-only Disk Parts3 VIDEO SPECIFICATIONS Version1.0 August 1996 (VI5-32 to VI5-37)
[0005]
[Problems to be solved by the invention]
Among the standards using the MPEG system, standards that can be recorded have been established, and in recording in those standards, the user can edit in frame units.
[0006]
As described above, as a characteristic of the MPEG system, at the time of encoding, a frame may have been compressed by being subjected to inter-frame prediction, so that a temporally backward frame in a GOP (or VOBU) may be used. For a frame, it takes time to decode itself. For example, in the case of decoding B6 in the GOP structure shown in FIG. 7, decoding is completed only after decoding a total of four frames including I2, P5, P8 and B6 itself.
[0007]
Then, considering the display time with reference to FIG. 9, for example, as shown in FIG. 8, the reproduction ends at P'5 of VOBU0, and the reproduction is restarted from B6 of VOBU1. If the editing is performed, when switching the frame from P'5 of VOBU0 to B6 of VOBU1, P'5 is output for three frame times before decoding of B6 of VOBU1 is completed. At this point, the same screen is repeatedly output, and seamless reproduction is not performed.
[0008]
Therefore, by using another decoding device, if the decoding of the data (up to I2, P5, P8) after the editing point is completed in advance at the time of reproduction at the editing point, the switching at the editing point can be performed. It will be done smoothly.
[0009]
However, when such a means is used, the cost is increased, the resource is inefficiently used due to over-spec, and the power consumption is increased.
Therefore, the present invention uses only one decoding device to decode data after an editing point in advance so that seamless playback is possible as much as possible in reproducing the editing point of the compressed encoded data. It is another object of the present invention to provide a coded data decoding device and a coded data decoding method capable of decoding coded data after an edit point by using the time remaining in decoding of the coded data before the edit point. .
[0010]
[Means for Solving the Problems]
In order to solve the above-mentioned problem, a coded data decoding apparatus according to claim 1 of the present invention is characterized in that coded data used for reproducing edited data by connecting coded image data by an arbitrary frame. The decoding device of
Decoding means for decoding the input encoded image data; storage means for storing the decoded data output from the decoding means for display and decoding of other encoded image data; Decoding time allocating means for controlling the time of the difference between the display time of the image encoded data before the edit point and the decoding time to be applied to the decoding of the image encoded data after the edit point reproduced temporally later; Stores up to which position in the frame the image-encoded data after the edit point, which will be reproduced later, has been decoded in the previous decoding, and starts decoding from the point when the previous decoding was completed in the next decoding Decoding position management means for controlling the decoding position.
[0011]
Further, the decoding device for encoded data according to claim 2 is the decoding device according to claim 1, wherein the image encoded data is in an MPEG format defined by ISO / IEC11172 or 13818, and is temporally later. The decoding position management unit of the image coded data after the editing point to be reproduced is in units of slices or macroblocks.
[0012]
Further, a decoding device for coded data according to claim 3 is the decoding device according to claim 1, further comprising means for counting the size of the decoded coded data at the time of decoding. The value is used to calculate the decoding position of the encoded image data reproduced later in time.
[0013]
A method for decoding encoded data according to claim 4 is a method for decoding encoded data when reproducing edited data by connecting image encoded data by an arbitrary frame,
A decoding step of decoding input image encoded data; a storage step of storing decoded data output from the decoding step for display and decoding of other image encoded data; A decoding time allocating step for controlling the difference between the display time and the decoding time of the image encoded data before the edit point to be applied to the decoding of the image encoded data after the edit point reproduced temporally later; Stores up to which position in the frame the image-encoded data after the edit point, which will be reproduced later, has been decoded in the previous decoding, and starts decoding from the point when the previous decoding was completed in the next decoding And a decoding position management step for controlling the decoding.
[0014]
According to a fifth aspect of the present invention, in the decoding method of the fourth aspect, in the decoding method of the fourth aspect, the coded image data is an MPEG format defined by ISO / IEC11172 or 13818, and is temporally later. This is a decoding method in which a decoding position management unit of image encoded data after an editing point to be reproduced is a slice or a macroblock unit.
[0015]
Further, the decoding method of the encoded data according to claim 6 is the decoding method according to claim 4, further comprising the step of counting the size of the decoded encoded data at the time of decoding. This is a decoding method for calculating a decoding position of image encoded data reproduced later in time using a value.
[0016]
According to the decoding device and the decoding method, in reproducing the edit point of the compressed encoded image data, the decoding of the encoded data after the edit point is the time remaining in the decoding of the encoded data before the edit point. By using this method in advance, seamless reproduction can be performed with only one decoding device.
[0017]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, an apparatus and method for decoding encoded data according to an embodiment of the present invention will be described with reference to the accompanying drawings, taking a DVD video recording standard as an example.
[0018]
The coded data decoding device according to the present embodiment is a device for connecting two compressed image coded data by an arbitrary frame to reproduce edited data, and includes at least a coded data management unit and , An encoded data decoding unit, and the encoded image data extracted by the encoded data management unit is input to the encoded data decoding unit to obtain predetermined output data.
[0019]
Although the order of the description is reversed, first, the encoded data decoding unit will be described with reference to FIG.
The coded data decoding unit decodes the coded image data (hereinafter, also simply referred to as coded data) extracted by the coded data management unit, and outputs the decoded data from the decoding unit 10. A storage unit 11 for storing the decoded data for displaying the decoded data and decoding other encoded data, and an edited encoded image obtained by connecting two compressed image encoded data by an arbitrary frame. In the case of reproducing data, the difference between the display time and the decoding time of the encoded data before the editing point reproduced earlier in time based on the external control signal from the navigation data analyzing means indicating the reproduction order and the like. Decoding time allocating means 12 for controlling decoding of encoded data after an editing point reproduced later in time, and a code after an editing point reproduced later in time. Decoding position management means 13 for storing the position of the frame to which the coded data has been decoded in the previous decoding, and controlling the decoding to start from the time when the previous decoding was completed at the time of the next decoding. ing.
[0020]
Next, the encoded data management unit will be described with reference to FIG.
The coded data management unit determines whether the input coded data is before the edit point or after the edit point based on the external control signal from the navigation data analysis unit in the same manner as above, and sorts the data. , A first coded data storage means 21 for storing the coded data before the edit point among those sorted by the separation means 20, and a coded data after the edit point as well. Based on control signals from the second encoded data storage means 22 and the decoding time allocating means 12, either one of the first encoded data storage means 21 and the second encoded data storage means 22 In addition to determining whether the encoded data is necessary, the encoded data read position is determined based on the control signal from the decoding position management means 13 to output the encoded data. A data reading means 23 is provided.
[0021]
In the above configuration, in order to realize seamless reproduction at the time of reproduction of the edit point, it is necessary to decode the encoded data after the edit point in advance.
For that purpose, it is necessary to first take the encoded data after the editing point into the decoding device in advance. As the transfer means, the size of the encoded data storage means 21 and 22 is secured to be larger than usual, and when the transfer of the encoded data before the edit point is completed, the encoded data after the edit point is completed. There are a method of starting the transfer and a method of transferring the data before the edit point and the encoded data after the edit point by time-division multiplexing, and in order to realize seamless reproduction at the edit point, the following (1) It is necessary to consider the times T1 and T2 shown in the equations and the equation (2).
[0022]
(Equation 1)
T1 = Vs × (Td × N) / (Vs−Td) + (M / R) (1)
T2 = {(Td × N) / (Vs−Td)} × Vs (2)
However, in the above formula, each symbol represents the following contents.
[0023]
Vs: 1 frame time Td: time required to decode one frame N: number of frames required to reach the output picture after the editing point M: maximum number of bits of one frame R: transfer bit rate, ie To ensure seamless playback at the edit point, advance of the encoded data after the edit point is started before the time T1 shown in the equation (1) before the edit point, and the time T2 shown in the equation (2) is started. The decoding of the encoded data after the editing point needs to be started earlier.
[0024]
Here, the decoding timing of the encoded data in the decoding device will be described with reference to FIG.
The editing points in this case are P'5 of VOBU0 and B6 of VOBU1 as shown in FIG. 8 described in the related art, and seamless reproduction at the editing points will be described based on FIG.
[0025]
First, the B'0 frame of VOBU0 before the editing point is decoded. Here, since the decoding apparatus always has the decoding capability of completing the decoding of one frame within at least one frame time, when the decoding of the frame of B'0 is completed, the decoding apparatus still has the decoding capability. There is a high possibility that there is enough time before decoding of the B′1 frame of VOBU0 starts. If there is enough time, decoding of the VOBU 1 after the editing point is started using that time. In this example, I2 of VOBU1 is decoded. During the decoding of I2 of VOBU1, when the time to start decoding the frame of B′1 of VOBU1 before the editing point (the decoding device needs to start decoding in units of one frame time) comes. The decoding position management unit 13 stores up to which slice of I2 of the VOBU 1 decoding has been completed, and further stores the pointer position of the slice in the encoded data storage unit 22. Next, the decoding of B'1 of VOBU0 is performed, and when the decoding of B'1 is completed, the decoding of the I2 of VOBU1 that has been completed halfway is performed, and the slice number and the encoded data storage unit of the slice are stored. Using the pointer position at 22, decoding of I2 is started halfway. Thereafter, similarly, decoding of I2, skipping of B0 and B1, decoding of P5, skipping of B3 and B4, and decoding of P8 are performed using the time remaining in decoding the encoded data before the editing point. In this way, by switching to the decoding of P8 of VOBU1 before the editing point in advance, the switching at the editing point can be performed seamlessly.
[0026]
Next, a control flow in the decoding time allocating means 12 in the encoded data decoding device will be described with reference to FIG.
First, when the process is started in step S400, the VOBU before the editing point is decoded in step S401. Next, in step S402, it is determined whether or not decoding of the VOBU after the editing point is necessary. If the reproduction of the VOBU after the editing point is not performed, the decoding of the VOBU after the editing point is already completed. If decoding of the VOBU is not necessary, the process ends in step S406.
[0027]
If it is determined in step S402 that the VOBU after the editing point needs to be decoded, it is determined in step S403 that the time (display time of the VOBU frame before the editing point−decoding time of the VOBU frame before the editing point) is left. It is determined whether or not there is no time left, and the process ends in step S406.
[0028]
If time remains in step 403, the VOBU after the editing point is decoded in step S404. Then, in step S405, it is determined whether or not there is time left (display time of the VOBU frame before the editing point−decoding time of the VOBU frame before the editing point−decoding time of the VOBU after the editing point). If not, the process ends in step S406.
[0029]
If there is time left in step S405, the flow returns to step S404 to decode the VOBU after the editing point, and thereafter repeats this processing until the time for decoding the VOBU before the next editing point comes.
[0030]
Next, a control flow in the preceding decoding step S404 of the data after the editing point in the encoded data decoding device will be described with reference to FIG.
First, in step S500, when the process is started, in step S501, the edit point reproduction start time indicating the edit point is compared with the decoded PTS (Presentation Time Stamp), and if the PTS is larger, the edit point Since it can be determined that the preceding decoding of the subsequent VOBU1 has reached the end point, the process ends in step S507.
[0031]
If the PTS is smaller in step S501, it is determined that it is necessary to perform the preceding decoding of the VOBU1 after the editing point, and it is determined in step S502 whether decoding is currently being performed. If it is not in the middle of decoding, it is determined in step S503 whether the next picture to be decoded is a B picture. If it is a B picture, there is no need to decode it. To skip the picture, and the process ends in step S507.
[0032]
If it is determined in step S503 that the next picture to be decoded is not a B picture, it is necessary to perform decoding. Therefore, decoding is performed in step S506. After decoding of the slice is completed, the process is completed in step S507.
[0033]
If it is determined in step S502 that the decoding is in progress, in step S504, the read pointer of the encoded data storage unit 22 of the image is returned to the next slice header of the slice for which decoding is completed. The decoding is performed in step S506, and after the decoding of the slice is completed, the process is completed in step S507.
[0034]
Next, the control flow of the decoding position management of the data after the editing point in the encoded data decoding device, that is, the control flow in the decoding position management means 13 will be described with reference to FIG.
[0035]
First, in step S600, when the process is started, in step S601, the detected slice number of the slice header is stored. Next, the pointer position in the encoded data storage unit 22 of the slice header detected in step S602 is stored. Here, in order to accurately grasp the position of the slice header, a means (mechanism) for counting the data size of the decoded coded data is provided during decoding, and the data size and the previous coded data storage means are provided. By using the pointer position at 22, the position is grasped. Finally, the slice is decoded in step S603, and after the decoding is completed, the process is completed in step S604.
[0036]
Here, the main steps of the decoding method are specified.
That is, the decoding method according to the present embodiment is a decoding method for reproducing edited data by connecting encoded image data by an arbitrary frame, and decodes input encoded image data. A decoding step, a storage step of storing the decoded data output from the decoding step for display and decoding of other encoded image data, and a step of storing the encoded image data before the edit point reproduced earlier in time. A decoding time allocating step of controlling the time of the difference between the display time and the decoding time so as to be applied to decoding of the image coded data after the editing point reproduced later in time; and after the editing point reproduced later in time. The decoding position that stores the position of the frame in which the image coded data has been decoded in the previous decoding, and controls the decoding to start from the time when the previous decoding was completed in the next decoding. A method and a management step,
The image coded data in the above decoding method is the MPEG system specified by ISO / IEC11172 or 13818, and the decoding position management unit of the image coded data after the edit point reproduced later in time is slice or macro. This is a block-based method,
The method further comprises the step of counting the size of the decoded data at the time of decoding in the above decoding method, and using the value of the counter to calculate the decoding position of the encoded image data reproduced later in time. is there.
[0037]
As described above, in the reproduction of the edit point of the compressed image encoded data, decoding of the encoded data after the edit point is performed in advance using the time remaining in decoding the encoded data before the edit point. By doing so, seamless reproduction can be performed with only one decoding device. Therefore, it is possible to prevent an increase in cost and an over-specification as compared with a case where two decoding devices are used.
[0038]
By the way, in the above-described embodiment, the preceding decoding of the data after the editing point has been described to be managed in units of slices. However, the management can be performed in units of frames, pictures, or macroblocks.
[0039]
In the above-described embodiment, the decoding device has been described in such a manner that the decoding device stores the position of the detected slice in the bit buffer of the video each time the slice header is detected by using the means for counting the size of the stream. However, the position of the bit buffer can be stored in units of frames, pictures, or macroblocks.
[0040]
Further, in the above-described embodiment, the completion confirmation of the decoding is managed by the PTS, but the completion of the decoding can be confirmed by using the Temporal Reference.
[0041]
In the above embodiment, the management of the pointer position in the encoded data storage means is realized by using the counter and the pointer. However, the management can be performed by using only the pointer. However, in this case, it is necessary to consider the remaining amounts of all buffers arranged between the encoded data storage unit and the decoding unit.
[0042]
Further, in the above-described embodiment, the management unit of the stream before and after the editing point is performed by VOBU, but may be performed by GOP.
[0043]
【The invention's effect】
As described above, according to the decoding device and the decoding method of the present invention, in the reproduction of the edit point of the image encoded data, the decoding of the encoded data after the edit point is performed by decoding the encoded data before the edit point. By performing the processing in advance by using the surplus time, seamless reproduction can be performed with only one decoding device.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a configuration example of an encoded data decoding unit in an encoded data decoding device according to an embodiment of the present invention.
FIG. 2 is a block diagram illustrating a configuration example of an encoded data management unit in the encoded data decoding device.
FIG. 3 is a diagram showing a decoding timing chart in the encoded data decoding device.
FIG. 4 is a diagram showing a control flow in decoding time allocating means in the encoded data decoding apparatus.
FIG. 5 is a diagram showing a control flow in preceding decoding of data after an editing point in the encoded data decoding apparatus.
FIG. 6 is a diagram showing a control flow in decoding position management of data after an edit point in the encoded data decoding apparatus.
FIG. 7 is a diagram showing an inter-frame structure in the MPEG system.
FIG. 8 is a diagram illustrating an example of a data structure of an edit point.
FIG. 9 is a diagram showing a decoding timing chart at an edit point.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 10 Decoding means 11 Storage means 12 Decoding time allocating means 13 Decoding position management means 20 Separation means 21 Encoded data storage means 22 Encoded data storage means 23 Encoded data read means

Claims (6)

A decoding device for reproducing edited data by connecting image encoded data with an arbitrary frame,
Decoding means for decoding the input image encoded data,
Storage means for storing the decoded data output from the decoding means for display and decoding of other image encoded data,
Control so that the difference between the display time and the decoding time of the encoded image data before the edit point reproduced earlier in time is used for decoding the encoded image data after the edit point reproduced later in time. Decoding time allocating means;
Stores up to which position in the frame the image-encoded data after the editing point, which is reproduced later in time, has been decoded in the previous decoding. An encoded data decoding apparatus, comprising: a decoding position management unit that controls to start the decoding. The image coded data is in the MPEG format defined by ISO / IEC11172 or 13818, and the decoding position management unit of the image coded data after the editing point reproduced later in time is a slice or macroblock unit. The decoding device for coded data according to claim 1, wherein: Means for counting the size of the decoded data at the time of decoding, and using the value of the counter to calculate the decoding position of the encoded image data reproduced later in time; An apparatus for decoding encoded data according to claim 1. A decoding method for reproducing edited data by connecting image encoded data by an arbitrary frame,
A decoding step of decoding the input image encoded data,
A storage step of storing the decoded data output from the decoding step for display and decoding of other encoded image data,
Control so that the difference between the display time and the decoding time of the encoded image data before the edit point reproduced earlier in time is used for decoding the encoded image data after the edit point reproduced later in time. A decoding time allocating step;
Stores up to which position in the frame the image-encoded data after the editing point, which is reproduced later in time, has been decoded in the previous decoding. A decoding position management step of controlling to start the decoding. The image coded data is in the MPEG format defined by ISO / IEC11172 or 13818, and the decoding position management unit of the image coded data after the editing point reproduced later in time is a slice or macroblock unit. The method for decoding encoded data according to claim 4, characterized in that: 5. The method according to claim 4, further comprising the step of counting the size of the decoded data at the time of decoding, and using the value of the counter to calculate a decoding position of the encoded image data to be reproduced later in time. 3. A method for decoding encoded data according to item 1.

Images