JP2000295567A - Coded data editor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a coded data editor by which an edited picture is edited so as to be displayed as intended without causing a decoded picture after the edit to be different depending on a type of a decoder. SOLUTION: In succession to 1st moving picture coded data of a program, 2nd moving picture coded data of 1 GOP outputted from an adder 14 and succeeding 2nd moving picture coded data are selected and composited in this order by switching. Thus, when a closed GOP flag of a 1st 1 GOP at an edit point is set to '0', a B picture in the GOP is copied to have the same picture contents as those of a just preceding picture in the GOP in the case of decoding.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a coded data editing apparatus, and more particularly to a coded data editing apparatus for editing coded moving picture data obtained by compressing and coding a moving picture signal.

[0002]

2. Description of the Related Art MPEG2 (Moving Picture Experts Groove) has been widely used as a moving picture coding method.
There is a video coding method called up phase 2). MP
In EG2, motion compensation (MC) prediction between images and discrete cosine transform (DCT) of 8 × 8 pixels are performed, and a signal obtained by this is further subjected to quantization and variable length coding. The types of MC prediction include forward prediction using a past image as a reference image, backward prediction using a future image as a reference image, bidirectional prediction using both past and future images as reference images, and prediction. There are no intra modes.

The MC prediction mode can be set for each macroblock of 16 × 16 pixels, but a mode that can be used is determined depending on the type (picture type) of a coded image (picture). There are three picture types. These are an I picture composed of only intra macroblocks, a P picture composed of intra and forward prediction macroblocks, and a B picture in which all MC prediction modes are allowed.

[0004] Here, the I picture is obtained by subjecting the original picture itself to DCT conversion, quantization and variable length coding without using prediction, so that it is possible to decode with a single piece of coded data. The P picture is obtained by adding an MC prediction error signal with a past, already coded I or P picture in the order of the input picture to D.
CT transformed, quantized, variable length coded, and B
The picture is obtained by DCT-transforming, quantizing, and variable-length coding the MC prediction error signal with the past and future I or P pictures already coded. Therefore, P and B
Before decoding a picture, it is necessary to decode a reference image starting from an I picture.

In MPEG2, a GOP (Group Of Pictures) composed of an arbitrary number of pictures of the above type
Can have a hierarchy of The first picture to be encoded in this GOP is defined as an I picture, and a GOP header indicating the beginning of the GOP is inserted before this I picture. The GOP header indicates whether the time code and the coded data constituting the GOP can be decoded only by the data in the GOP, that is, whether the GOP is an independent GOP (closed GOP) that does not refer to the image data of the previous GOP. Flag (closed_gop),
Although it is originally necessary to refer to the image data of the previous GOP, there is a flag (broken_link) indicating that the image data cannot be edited, so that the editing can be performed in GOP units.

FIG. 6 is a block diagram showing the configuration of a conventional coded data editing apparatus provided with such a coded data supply unit based on MPEG2. FIG. 7 shows an example of encoded data to be edited. Editing is performed in GOP units. In FIG. 6, the moving picture coded data supply unit 1
In (a), the part before the edit point (edit point) 9 is 7
a, a first encoded video stream according to MPEG2, the latter of which is schematically indicated by 7b, is generated and supplied to the terminal 3a of the switch 3.

[0007] Also, the moving picture coded data supply unit 2 is configured as shown in FIG.
In (b), the part before the edit point (edit point) 9 is 8
a, a second encoded video stream according to MPEG2, the latter of which is schematically indicated by 8b, is generated and supplied to the terminal 3b of the switch 3. It should be noted that here, the first
It is assumed that the coding rates of the moving picture coded stream and the second moving picture coded stream are fixed and equal.

When the switch 3 reaches the editing point 9, the switch 3 is switched from the terminal 3a to the terminal 3b. As a result, the movable terminal 3c of the switch 3 receives the signal shown in FIG.
As schematically shown in FIG. 2, a first encoded video stream 7a before the edit point (edit point) 9 and a second encoded video stream 8b after the edit point 9 are time-divided. One multiplexed moving picture encoded data is extracted.

[0009]

However, when a part of the coded video data stream is edited as described above, the broken link (broken_link) is included in the GOP header of the coded video data stream immediately after editing. By setting the flag, there is no problem in the syntax of MPEG2, but at the time of decoding after editing, depending on the decoder, the first I in the encoded video data immediately after editing may be used.
If a B picture following a picture exists before editing, but a picture in the immediately preceding GOP that does not exist after editing is used as a reference image, the B picture cannot be correctly decoded.

In addition, since the operation of the decoder of the GOP with the broken link flag set has not been specified, the reproduced image at this time differs depending on the configuration of the decoder.

The present invention has been made in view of the above points,
It is an object of the present invention to provide an encoded data editing apparatus capable of editing an edited image so that the decoded image is displayed as intended without the decoded image being changed depending on the type of decoder.

[0012]

According to a first aspect of the present invention, there is provided a moving image encoding apparatus comprising: a first moving picture encoding apparatus comprising: a frame using predictive coding; and a frame not using predictive coding. In a coded data editing device for performing an edit by connecting second coded video data consisting of a frame using predictive coding and a frame not using predictive coding at an editing point, the second video coding Detecting means for detecting a frame using predictive encoding from the second encoded video data, wherein the frame using non-existent data is used as a reference frame; and a frame using predictive encoding detected by the detecting means. And re-encoding means for re-encoding a frame that does not use predictive encoding and has the same image content as the immediately preceding frame.

Further, in order to achieve the above object, the second invention is based on the MPEG system, and comprises an I picture composed of intra macroblocks obtained by compressing and encoding an original image itself without using prediction, and an intra picture. Among the B pictures obtained by compressing and encoding the prediction error signal between the P picture composed of the macroblock and the forward prediction macroblock and the past and future already encoded I picture or P picture, one I picture is used. MPEG encoded first moving image data in GOP units including a plurality of P pictures and B pictures
In a coded data editing apparatus that performs editing by connecting at the edit point the second coded video data of the GOP unit that has been compression-coded according to the system, an edit point immediately after the edit point of the second coded video data is used. The first GOP is the previous GOP
A first determination unit that determines whether the GOP is an independent GOP that does not refer to
The first GOP of the moving picture encoded data is independent GOP
When it is determined that the B-picture is not a reference picture, a second determining unit that detects a B-picture that does not use a picture in the GOP as a reference image from the pictures in the GOP, and a B-picture input from the second determining unit A changing unit that changes a value indicating a picture to be used as a reference image at the time of decoding to a value for copying an image of a picture immediately before the B picture and outputs the changed value. The first moving image encoded data immediately after the edit point is determined by each picture other than the B picture detected by the determining unit, or by each picture of the GOP determined to be an independent GOP by the first determining unit. Of a GOP.

In the first and second aspects of the present invention, the second moving image encoded data immediately after the editing point
Since a picture (frame) having a frame that does not exist in the moving image encoded data as a reference frame is re-encoded into a frame that does not use predictive encoding and has the same image content as the immediately preceding frame, An image can be decoded even in a frame (picture) in which a reference frame no longer exists due to editing.

[0015]

Next, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of an encoded data editing apparatus according to the present invention. In the figure, the input coded video data is supplied to a GOP header determination unit 11. This input video encoded data is
This is encoded data obtained by compressing and encoding a moving image signal by the above-described MPEG2, and is a second data connected after the first moving image encoded data of a certain program as shown in FIG. Of the first encoded GOP immediately after the edit point of the encoded video data of
Is not performed.

Here, this input moving picture coded data, as schematically shown in, for example, FIG. 5A, forms one GOP with 12 pictures, and each picture in one GOP has a leading picture. This is so-called M = 3, N = 12 encoded data in which two B pictures and one P picture are alternately and repeatedly arranged in an I picture. In addition, as shown in FIG.
P constitutes the moving picture coded data of one program shown in FIG.

The GOP header judging unit 11 determines whether the previous GOP in the GOP header of the input moving picture encoded data having the above configuration
It is determined whether or not a closed GOP (closed_gop) flag indicating whether or not the GOP is an independent GOP that does not refer to the image data is “0”. When the closed GOP flag is “0”, it is not an independent GOP. ,
Since it indicates that decoding is not possible with only the encoded data in the GOP, each picture (each frame) of the GOP is supplied to the re-encoded frame determination unit 12, while the closed GOP flag is “1”. In some cases, the data is output as it is without passing through the re-encoded frame determination unit 12 and each circuit described later. This is because the decoding can be performed only with the encoded data in the GOP.

When the closed GOP flag is "0", the GOP header determining section 11
After changing the OP flag to “1”, each picture (frame) of the GOP is supplied to the re-encoded frame determination unit 12.

The re-coded frame determination section 12 determines a re-coded frame according to the flowchart shown in FIG. That is, it is determined from the picture type of each picture of the input GOP whether or not the input picture is a B picture (step 21). If the input picture is a B picture, the B picture is referred to a frame in the previous GOP. It is determined whether or not the B picture is present (step 22), and if so, the B picture is supplied to the re-encoding unit 13 for re-encoding (step 23).

On the other hand, when the input picture is determined to be an I-picture or a P-picture in step 21, or when the input picture is a B-picture but in step 22, the re-coded frame determination unit 12 refers to the frame in the previous GOP. If it is determined that the picture is not a B-picture,
The data is output as it is without being re-encoded by the re-encoding unit 13 (step 24). Thus, the closed GOP flag is “0” in the re-encoding unit 13,
Thereafter, only the B picture in one GOP changed to “1” is input.

As shown in FIG. 3, the re-encoding unit 13 includes an encoded frame memory 31 and an encoded MB type changing unit 3.
2 and a coded frame memory 33. The B picture input to the re-encoding unit 13 is supplied to the encoded frame memory 31 and temporarily stored, and then the macro block (MB) in the picture (frame) is encoded for each MB unit. It is supplied to the MB type changing unit 32.

The coded MB type changing unit 32 changes the MB according to the flowchart shown in FIG. 4 and supplies the processed MB to the coded frame memory 33. That is, the encoded MB
The type change unit 32 first determines that the input MB is the first M of the slice.
B is determined based on the presence or absence of a slice start code (step 41). Here, the macroblock (MB) is a basic unit of encoding in which a block of 8 pixels in the horizontal direction and 8 pixels in the vertical direction is composed of four luminance signals and two types of color difference signals, one each for a total of six, A slice is formed by collecting an arbitrary number of macro blocks. Further, the picture is divided by a plurality of slices of arbitrary length.

If it is determined in step 41 that the input MB is the first MB, the prediction mode is set to forward prediction, the value of the forward motion vector MVf is set to (0, 0), and coding is not required. (Step 42). This allows
The first MB has the same information content as the corresponding MB of the I picture in the same GOP at the time of decoding.

On the other hand, in step 41, the first MB of the slice
If it is determined that the MB is not motion compensated (M
C: No Motion Compensation, no coding is required, and a so-called skip MB with a code amount of zero is set (step 43). MC refers to detecting a motion vector by performing pattern matching on a motion area for each MB, shifting the motion vector by the amount of motion, and predicting the motion vector. The skip MB in step 43 has the same information content as the corresponding MB of the I picture in the same GOP at the time of decoding.

As a result, in each of the steps 42 and 43, when the MB is decoded, the MB has the same information content as the corresponding MB of the I picture in the same GOP (ie,
Copied), but the M at the beginning of the slice
In order to comply with the MPEG rule that B is required to be encoded and not to be a skip MB, different processing is performed between the first MB and the other MBs.

The B picture in one GOP whose closed GOP flag processed according to the flowchart of FIG. 4 is “0” and subsequently changed to “1” is shown in FIG.
Is temporarily stored in the coded frame memory 33, and is output when all MBs are collected, taken out via the adder 14 in FIG. Is connected immediately after

That is, the first 1 GOP immediately after the edit point of the first encoded video data of a certain program and the second encoded video data output from the adder 14
And the second and subsequent GOPs immediately after the edit point of the second moving picture encoded data that does not pass through the circuit of FIG. 1 are input to a switch circuit (not shown), and the first moving picture encoding is performed. Following the data, 1G output from the adder 14
OP second video encoded data and second and subsequent GOs
Switching and combining are performed in the order of the P second moving image encoded data.

Thus, the first one of the edit points is
When the closed GOP flag of the GOP is “0”, the B picture that uses a frame that does not exist in the GOP as a reference frame has the same GOP during decoding.
Since the image content is copied to the same image content as that of the I picture, the conventional disadvantage that the displayed image differs depending on the type of decoder can be eliminated.

[0029]

As described above, according to the present invention,
From the second video encoded data immediately after the edit point,
By re-encoding a picture (frame) having a frame that does not exist in the second encoded video data as a reference frame into a frame that does not use predictive encoding and has the same image content as the immediately preceding frame, Since the image can be decoded even in a frame (picture) in which the reference frame is no longer present due to the editing, it is possible to prevent the possibility that the displayed image differs depending on the type of the decoder. It can be obtained as the intended image at the time of editing.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of the present invention.

FIG. 2 is a flowchart for explaining the operation of a re-encoded frame determination unit in FIG. 1;

FIG. 3 is a block diagram of an embodiment of a re-encoding unit in FIG. 1;

FIG. 4 is a flowchart for explaining the operation of FIG. 3;

FIG. 5 is an explanatory diagram of a relationship between a picture and a GOP.

FIG. 6 is a block diagram of an example of a conventional device.

FIG. 7 is a schematic diagram illustrating the operation of FIG. 6;

[Explanation of symbols]

11 GOP header determination unit (detection unit, first determination unit) 12 re-encoded frame determination unit (detection unit, second determination unit) 13 re-encoding unit (re-encoding unit, change unit) 31, 33 code Frame memory 32 Encoding MB type change unit

Claims (2)

[Claims] The first moving image encoded data including a frame using predictive coding and a frame not using predictive coding includes a frame using predictive coding and a frame not using predictive coding. An encoded data editing apparatus for performing editing by connecting second encoded video data at an editing point, wherein a frame that does not exist in the second encoded video data is used as a reference frame. Detecting means for detecting a frame that has been detected from the second moving image encoded data, and predictive coding for setting a frame using predictive coding detected by the detecting means to have the same image content as the immediately preceding frame. Recoding means for recoding a frame not to be used. 2. An I picture composed of intra macroblocks obtained by compressing and coding an original image itself without using prediction according to the MPEG system, and a P picture composed of said intra macroblocks and forward predicted macroblocks When,
Among B pictures obtained by compression-encoding a prediction error signal of a past and future already encoded I picture or P picture, one I picture and another P picture and a plurality of B pictures are GOP units. An encoded data editing apparatus for performing an edit by connecting, at an edit point, second encoded video data of the GOP unit, which is compressed and encoded in accordance with the MPEG method, to first encoded video data, The first GOP immediately after the edit point of the second encoded video data is an independent GOP that does not refer to the previous GOP.
A first judging unit for judging from the header whether or not it is an OP; and when the first judging unit judges that the first GOP of the second encoded video data is not the independent GOP. A second determining unit that detects a B picture that does not use a picture in the GOP as a reference image from each picture in the GOP; and a reference image for decoding the B picture input from the second determining unit. A changing unit for changing a value indicating a picture to be used to a value for copying an image of a picture immediately before the B picture and outputting the copied image, wherein the B picture output from the changing unit and the second determination unit The position of the edit point is determined by each picture other than the detected B picture or by each picture of the GOP determined to be the independent GOP by the first determination unit. Coded data editing apparatus characterized by constituting the first GOP of the second encoded video data stream.