JP2001157161A - Device and method for editing image and program recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent the buffer of a decoder from being damaged due to the continuity of code quantity not being maintained when optional compressed data are pasted, when editing data that have been subjected to compression and encoding the data without decoding all the data. SOLUTION: This image-editing method decodes the section between a time T1 and a time T2 in an inputted prescribed compressed bit stream and adjusts code quantity, so as to make a VBV value be prescribed V2 at the time T2 in the case of decoding a compressed bit stream obtained by executing reencoding again, when a decoded image is encoded with 1st editing performed or without performing the editing, and presupposes the VBV value at the time T1 is Vx. The V2 is the VBV value, at the time T2 when the inputted compressed bit stream is decoded as is.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image editing method for editing digital AV data, an image editing apparatus, and a program recording medium.

[0002]

2. Description of the Related Art Conventionally, when editing image data compressed in the MPEG2 format, first, all the compressed image data is decoded, then the decoded image is edited, and the edited image is edited. Has been compressed and encoded again in the MPEG2 format.

For example, a case will be described in which a telop is inserted in an image from 100 frames to 250 frames counted from the head of compressed AV data. In this case, if there are a total of 3000 frames of compressed AV data, the AV data of all 3000 frames is decoded. Then, 100 of the decoded images are counted from the top.
The telop is inserted into the images from the frame to the 250th frame. When the insertion of the telop is completed, the AV data of all 3000 frames is compression-encoded again in the MPEG2 format.

[0004]

However, in the case of editing compressed data, all the data is decoded, then edited, and compressed and encoded again. Therefore, in the process of decoding and re-encoding. Image quality degradation cannot be avoided.

[0005] Further, when editing without decoding all the data that has been compression-encoded, it is difficult to separate the compressed data at an arbitrary time from the compressed data that has been variable-length encoded.

In the case of editing without decoding all of the data which has been compression-encoded, the compression data which has been inter-frame-encoded depends on the preceding and succeeding frames. It is difficult to separate by time.

[0007] When editing compressed data without decoding all the data, the code amount of the compressed data is controlled with an emphasis on continuity. Is not held, so that the decoder buffer is broken.

According to the present invention, when the compressed data is edited, the decoding is performed after all the data is decoded, and the compression encoding is performed again. Therefore, the image quality is deteriorated in the decoding-re-encoding process. In view of the above problem, it is an object of the present invention to provide an image editing method, an image editing apparatus, and a program recording medium in which even if compressed data is edited, the image quality is hardly deteriorated.

Further, according to the present invention, when editing without decoding all compression-encoded data, it is difficult to separate compressed data at an arbitrary time from variable-length-encoded compressed data. In view of the above problem, an object of the present invention is to provide an image editing method, an image editing device, and a program recording medium that can easily separate compressed data that has been subjected to variable-length encoding at an arbitrary time. .

Further, according to the present invention, it is difficult to separate compressed data that has been inter-frame encoded at an arbitrary time when editing without decoding all the data that has been compressed and encoded. An object of the present invention is to provide an image editing method, an image editing apparatus, and a program recording medium that can easily separate compressed data that has been inter-frame coded at an arbitrary time in consideration of the problem.

Further, according to the present invention, when editing without decoding all of the data which has been compression-encoded, if any compressed data is cut, the continuity of the code amount is not maintained, so that the buffer of the decoder is broken. Considering the problem,
It is an object of the present invention to provide an image editing method, an image editing apparatus, and a program recording medium that do not break a buffer of a decoder even if any compressed data is cut.

[0012]

In order to solve the above-mentioned problems, a first aspect of the present invention (corresponding to claim 1) comprises an input means for reading a compressed bit stream, and an input means for reading the compressed bit stream. When decoding the section from time T1 to T2 and encoding the decoded image with or without the first editing, the compressed bit stream obtained by performing the encoding is re-input. When decoding, at the time T2, the Video Buff
editing adjusting means for adjusting the code amount so that the value of the erring Verifyer (hereinafter referred to as VBV value) becomes a predetermined V2; and output means for outputting the compressed bit stream output by the editing adjusting means. The said T
Assuming that the VBV value at one time is Vx,
The V2 is a VBV at the time T2 when the input compressed bit stream is decoded as it is.
An image editing apparatus characterized in that the value is a value.

Further, a second aspect of the present invention (corresponding to claim 2)
Is an image editing apparatus according to the first aspect of the present invention, wherein the first editing is performed over all or only a part of the time from the time T1 to the time T2.

A third aspect of the present invention (corresponding to claim 3)
Means that Vx is the VB at the end of another compressed bitstream to be merged into the compressed bitstream
The image editing apparatus according to the first aspect of the present invention, wherein the image editing apparatus is a V value, that is, a VBV value at the end when the another compressed bit stream is decoded as it is.

A fourth aspect of the present invention (corresponding to claim 4)
When the edit adjustment unit performs the first edit between the time T1 and the time T2, the Vx is a VBV value of the input compressed bit stream at the time T1. VB when the input compressed bit stream is decoded until the time T1
An image editing apparatus according to the first aspect, wherein the image editing apparatus has a V value.

A fifth aspect of the present invention (corresponding to claim 5)
The editing adjustment means decodes the input compressed bit stream for a predetermined section between the T0 time and the T1 time before the T1 time, and 2. The method according to the first aspect of the present invention, wherein the VxV is a VBV value at the time T1 when the compressed bit stream after the second editing is decoded. An image editing device.

The sixth invention (corresponding to claim 6)
The second edit is a transition edit in which the video of the other compressed bit stream is gradually switched to the video of the input compressed bit stream.
The data before the time does not use the part before the T0 time of the input compressed bit stream, but instead uses the part of the other compressed bit stream before the part used for the transition editing. An image editing apparatus according to a fifth aspect of the present invention, characterized in that:

The seventh invention (corresponding to claim 7)
The image according to the second aspect of the present invention, wherein the first editing is performed at a part of the time from the T1 time to the T2 time, and the decoding is performed on a part related to the editing. Editing device.

An eighth aspect of the present invention (corresponding to claim 8)
The first editing is to insert a telop, and the decoding part is in units of macroblocks.
An image editing apparatus according to a seventh aspect of the present invention, characterized in that not only a macroblock into which a telop is inserted but also a macroblock referencing the macroblock as a motion vector is decoded.

A ninth aspect of the present invention (corresponding to claim 9)
In the section for adjusting the VBV value, when re-encoding,
The image editing apparatus according to any one of the first to eighth aspects of the present invention, characterized in that re-encoding is performed using the encoding parameters before decoding as they are.

In a tenth aspect of the present invention (corresponding to claim 10), the adjustment of the code amount is performed by adjusting a quantization level. Image editing device.

According to an eleventh aspect of the present invention (corresponding to claim 11), the code amount is adjusted by selecting a DCT coefficient. The image editing apparatus described in the above.

In a twelfth aspect of the present invention (corresponding to claim 12), the adjustment of the code amount is performed using dummy data. Image editing device.

According to a thirteenth aspect of the present invention (corresponding to claim 13), the V2 is replaced by the VBV value at the time T2, and the V2 is a predetermined time before or after the time T2.
An image editing apparatus according to any one of the first to twelfth aspects of the present invention, which may be a BV value.

According to a fourteenth aspect of the present invention (corresponding to claim 14), the Vx is a VBV value at the end when the another compressed bit stream is decoded. An image editing apparatus according to a third aspect of the present invention, wherein the VBV value before or after a predetermined amount may be used.

According to a fifteenth aspect of the present invention (corresponding to claim 15), the Vx is input instead of the VBV value at the time T1 of the input compressed bit stream. VBV before or after a predetermined amount from the T1 time of the compressed bit stream
An image editing apparatus according to a fourth aspect of the present invention, wherein the image editing apparatus may be a value.

According to a sixteenth aspect of the present invention (corresponding to claim 16), the Vx is a VBV value at the time T1 when the second edited and encoded data is decoded. An image editing apparatus according to a fifth aspect of the present invention, wherein the VBV value before the T1 time by a predetermined amount when the second post-encoded data is decoded may be used.

According to a seventeenth aspect of the present invention (corresponding to claim 17), input means for reading a compressed bit stream comprises:
When merging the read first compressed bit stream and the second compressed bit stream, when merging the first half of the first compressed bit stream and the second half of the second compressed bit stream, If the first half of the compressed bit stream has a frame that depends on the second half of the first compressed bit stream, the second half of the first compressed bit stream is used to determine the frame that depends on the second half. Decoding and re-encoding, if the second half of the second compressed bit stream has a frame that depends on the first half of the second compressed bit stream, use the first half of the second compressed bit stream Editing means for decoding and re-encoding the frame dependent on the first half, the combined compressed bit stream An image editing apparatus is characterized in that an output means for outputting the over arm.

According to an eighteenth aspect of the present invention (corresponding to claim 18), the section from time T1 to time T2 in the input predetermined compressed bit stream is decoded, and the decoded image is subjected to the first editing. When the compressed bit stream obtained by performing the encoding is again decoded when performing the encoding without performing or performing the editing, the encoding is performed so that the VBV value becomes a predetermined V2 at the time T2. An image editing method for adjusting the amount, on the assumption that the VBV value at the time T1 is Vx,
An image editing method, characterized in that it is a VBV value at the time T2 when the input compressed bit stream is decoded as it is.

According to a nineteenth aspect of the present invention (corresponding to claim 19), when the first compressed bit stream and the second compressed bit stream are combined, the first half of the first compressed bit stream and the second When combining the second half of the compressed bit stream of the second compressed bit stream, the first half of the first compressed bit stream includes a frame that depends on the second half of the first compressed bit stream.
Using the latter half of the compressed bit stream, decoding and re-encoding the frame dependent on the latter half,
The second part of the second compressed bit stream has the second
When there is a frame that depends on the first half of the compressed bit stream, the first frame of the second compressed bit stream is used to decode and re-encode the frame that depends on the first half. Image editing method.

According to a twentieth aspect of the present invention (corresponding to claim 20), all or some of the functions of all or some of the means of the image editing apparatus according to any of the first to seventeenth aspects of the present invention are provided. A program recording medium on which a program and / or data to be executed by a computer is recorded, the program recording medium being readable by a computer.

According to a twenty-first aspect of the present invention (corresponding to claim 21), all or a part of the operations of all or some of the steps of the image editing method according to the eighteenth or nineteenth aspect of the present invention. A program recording medium recording a program and / or data to be executed by a computer, wherein the program recording medium is readable by a computer.

[0033]

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

(First Embodiment) In this embodiment,
Two compressed streams in the form of an MPEG2 transport stream are read from the HDD 3, the latter half of the first compressed stream is cut, the first half of the second compressed stream is cut, and the remaining parts are combined to form one compressed stream. The case of creation will be described.

FIG. 1 shows a block diagram of an image editing apparatus according to the present embodiment.

The image editing apparatus 1 includes a GUI 7, an editing setting unit 8, an editing section reading unit 9, an IEEE1394I /
F10, decoding means 11, editing means 12, D / A conversion 1
3. It comprises an encoding unit 14 and a synthesizing unit 15.

The image editing apparatus 1 includes an IEEE 13
The 94 bus 5 and the monitor 6 are connected.

The IEEE 1394 bus 5 has an ST
B2, HDD3, and VCR4 are connected.

The IEEE 1394 bus 5 is an IEEE 1394 bus.
It is a bus conforming to 94-1995.

The GUI 7 is a graphics user interface for deciding what kind of editing is to be performed on the data which has been compression-encoded, and which part is to be edited.

The editing setting means 8 is a means for setting what kind of editing is to be performed on the data which has been MPEG2 compressed and encoded, and which part is to be edited by operating the GUI 7.

The editing section reading means 9 is a means for reading data of a section to be edited among the data which has been MPEG2 compression-encoded.

The IEEE 1394 I / F10 is an IEEE 1394 interface.
STB2, HDD3, VCR via 1394 bus 5
IEEE 1394 for exchanging commands and data with H.4
-1995 compliant interface.

The decoding means 11 is means for decoding data in a section to be edited.

The editing means 12 is a means for editing the decoded data of the section to be edited.

The D / A converter 13 is means for converting a GUI screen created by the GUI 7 or edited data into an analog signal.

The encoding means 14 is means for re-encoding a part of the data decoded for editing while adjusting the code amount.

The synthesizing means 15 is means for combining the data of the edited section and the data of other parts.

Further, as shown in FIG.
Is the demultiplexer 16, the transport buffer 1
7a to e, multiplex buffers 18a to 18e, elementary buffers 19a to e, decoders 20a to e, rearrangement buffers 21a to e, and changeover switches 22a to 22e.

The demultiplexer 16 is means for separating a transport stream according to a PID attached to a transport packet of an MPEG2 transport stream.

The transport buffers 17a to 17e are 5
Demultiplexer 1 has a size of 12 bytes.
Means for buffering the bursty, high-speed data stream separated in 6. Multiplex buffers 18a-
e is a means for buffering the multiplex jitter of 4 ms and the PES overhead so that the data transfer speed to the next-stage elementary stream buffers 19a to 19e can be consistent.

Elementary stream buffer 19a
To e are VBVs (Video Buffers to be described later).
ng Verifier) buffer.

The decoders 20a to 20e are means for expanding the elementary stream.

The rearrangement buffers 21a to 21e are buffers for delaying the display of the I frame or the P frame. The changeover switches 22a to 22e are means for rearranging and outputting the decoded frames in the order in which they are displayed.

As shown in FIG. 3, the decoder 20a
Ｅe are respectively composed of a variable length decoding 23, an inverse quantization 24, an inverse DCT 25, an adder 26, a format conversion 27, a video memory 28, and a motion compensation prediction 29.

The variable length decoder 23 is means for decoding macroblock coded information and separating the coding mode, motion vector, quantization information and quantized DCT coefficients.

The inverse quantization 24 is means for restoring DCT coefficients from the decoded 8 × 8 quantized DCT coefficients.

The inverse DCT 25 is means for performing an inverse DCT transform and converting the data into pixel space data.

The adder 26 is a means for outputting as it is in the intra coding mode, and adding and outputting the block data subjected to the motion compensation prediction in the motion compensation prediction mode.

The format converter 27 is means for converting the decoded data into a spatial resolution for processing.

The video memory 28 is a memory for temporarily storing the I and P frames because they need to be used as reference images in the subsequent decoding processing.

The motion compensation prediction 29 is means for performing motion compensation prediction from a prediction mode and a motion vector.

Further, as shown in FIG.
Are format conversion 30, screen rearrangement 31, adder 32, DCT 33, quantization 34, variable length coding 35, buffer 36, inverse quantization 37, inverse DCT 38, adder 3
9, video memory 40, motion compensation prediction 41, changeover switch 42, rate control 43, invalid data insertion 44, DCT
It consists of coefficient removal 45.

The format converter 30 is a means for converting into a spatial resolution used for encoding.

The screen rearrangement 31 is means for rearranging screens according to I, B, and P frames.

The adder 32 calculates the difference between the macroblock video data and the macroblock video data obtained by motion prediction from the reference screen in the case of the motion compensation prediction mode, and converts the macroblock video data in the case of the intra coding mode. Output means.

The DCT 33 is a means for performing a DCT transform on a block basis of 8 pixels × 8 lines to convert to a spatial frequency domain.

The quantization 34 is means for quantizing the converted 8 × 8 DCT coefficients according to target bits and visual characteristics, and sequentially scanning from low frequency components to convert to one-dimensional information.

The variable length coding 35 includes macroblock coding information such as a coding mode and a motion vector and quantization DC
This is a means for encoding the T coefficient by a variable length code that assigns a shorter code to data having a higher appearance frequency.

The buffer 36 is a buffer for temporarily storing the variable-length coded data.

The inverse quantization 37 is a means for inversely quantizing a quantized DCT coefficient of an I or P frame.

The inverse DCT 38 is a means for performing an inverse DCT transform on the 8 × 8 inversely quantized DCT coefficients to convert them into pixel space data.

The video memory 40 is means for holding the converted pixel space data as a reference screen.

The motion compensation prediction 41 is means for performing motion prediction from a reference screen.

The rate control 43 is means for allocating bits to each frame, performing rate control using a VBV buffer described later, and performing adaptive quantization using visual characteristics.

The invalid data insertion 44 is means for inserting invalid data into the encoded data to increase the code amount.

The DCT coefficient removal 45 is means for removing the high frequency component of the DCT coefficient of the coded data to reduce the code amount.

The HDD 3 is a hard disk that stores data that has been MPEG2 compressed and encoded. VCR4
Is a video cassette recorder for recording MPEG2 compression-encoded data.

The STB 2 is a means for receiving a broadcast wave transmitted from a broadcasting station and displaying MPEG2 compression-encoded data recorded on the HDD 3 or VCR 4. As shown in FIG. 5, the STB 2 is a tuner 4
7, descrambler 48, IEEE1394 I / F4
9, the decoding means 50 and the D / A converter 13.
The STB 2 also includes an antenna 46, a monitor 6, an IEEE
The E1394 bus 5 is connected.

The tuner 47 is means for receiving and demodulating a broadcast wave transmitted from a broadcast station. The descrambler 48 is means for decoding the scramble of the demodulated data. The IEEE 1394 I / F 49 is an IEEE 1394-1995 compliant interface for exchanging data and commands with the image editing apparatus 1, HDD 3, and VCR 4. The decoding unit 50 is a unit that decodes the MPEG2 transport stream. D / A conversion 1
A unit 3 converts the decoded data into an analog signal.

As shown in FIG. 6, the decoding means 50 comprises a demultiplexer 16, a transport buffer 17f,
It comprises a multiplex buffer 18f, an elementary stream buffer 19f, a decoder 20f, a rearrangement buffer 21f, and a changeover switch 22f.

The demultiplexer 16 is means for separating a transport stream according to a PID attached to a transport packet of an MPEG2 transport stream. The transport buffer 17f
It has a size of 512 bytes and is a means for buffering a bursty high-speed data stream separated by the demultiplexer 16. Multiplex buffer 18f
Is means for buffering 4 ms of multiplexed jitter and PES overhead so that the data transfer speed to the next-stage elementary stream buffers 19a to 19e can be consistent. The elementary stream buffer 19f is a VBV (Video Bufferin).
g Verify) buffer.

The decoder 20f is a means for expanding the elementary stream.

The rearrangement buffer 21f is a buffer for delaying the display of the I frame or the P frame.
The changeover switch 22f is a unit that rearranges and outputs the decoded frames in the order in which they are displayed.

Also, as shown in FIG. 3 described above, the decoder 20f performs variable length decoding 23 and inverse quantization 2 respectively.
4, inverse DCT 25, adder 26, format conversion 2
7, a video memory 28, and a motion compensation prediction 29.

The antenna 46 is means for converting a broadcast wave into an electric signal. The monitor 6 is a means for displaying an image of an analog signal.

Next, the operation of this embodiment will be described.

In this embodiment, as shown in FIG. 8, the image editing apparatus 1 combines the first half of the compressed stream A60 in the form of the MPEG2 transport stream and the second half of the compressed stream B61 into one compressed stream. C
The case of generating 62 will be described.

For this purpose, the edit setting means 8 sets the GUI 7
First, the user sets an object to be edited while looking at the GUI screen created in step. That is, a compressed stream to be read is specified. This designation is performed by designating a device to be read and a file name of data stored in the device.

Now, it is assumed that the HDD 3 is specified as a device to be read, and the compressed stream A60 and the compressed stream B61 are specified as file names.

Then, the editing setting means 8 instructs the editing section reading means 9 to read the compressed stream A60 and the compressed stream B61 from the HDD 3. The editing section reading means 9 transmits the compressed stream A60 and the compressed stream B61 via the IEEE1394 I / F10.
Is read from the HDD 3. The read data is passed from the editing setting unit 8 to the decoding unit 11 and is decoded. G
The UI 7 creates a GUI screen using the decrypted data. The D / A converter 13 converts this screen into an analog signal, and the monitor 6 displays the analog signal.

The edit setting means 8 sets how to edit the compressed stream A60 and the compressed stream B61 with reference to the GUI screen. Now, as shown in FIG. 8, the first 100 frames from the beginning of the compressed stream A60 are left, and the 101st frame and thereafter are removed. The previous part is deleted, leaving the following part as the boundary of the part. Then, it is assumed that a process of generating one compressed stream C62 by combining the end of the portion from the beginning to the 100th frame of the compressed stream A60 with the portion after the time T1 of the compressed frame B61.

The editing setting means 9 outputs the compressed stream B61
Is instructed to read from the portion after the time T1 to the portion at the time T2 after two I frames have appeared.

In response to this, the editing section reading means 9
The section from the time T1 to the time T2 of the compressed stream B61 is read and output to the decoding means 11.

As shown in FIG. 2, the demultiplexer 16 of the decoding means 11 outputs the read compressed stream B6
1 by identifying the PID. The compressed stream B61 is output to the transport buffer 17a.

When the transport buffer of the compressed stream B61 is buffered, the transport buffer 17a outputs the transport packet to the multiplex buffer 18a at a constant transfer rate. The multiplexing buffer 18a buffers the 4 ms multiplexing jitter and the overhead of the elementary stream, and functions so that the data transfer speed to the next-stage elementary stream buffer is consistent. The elementary stream buffer 19a supplies one frame of data to the decoder 20a.

As shown in FIG. 3, when one frame of data is transmitted from the elementary stream buffer 19a, the variable length decoding 23 of the decoder 20 decodes the macroblock coded information, , Motion vectors, quantization information and quantized DCT coefficients. Next, the inverse quantization 24 performs the decoded 8 × 8 quantized DCT.
Restore the DCT coefficients from the coefficients. The inverse DCT 25 is the inverse D
A CT conversion is performed to convert the data into pixel space data. The adder 26 outputs the data as it is in the intra-encoding mode, and adds and outputs the block data subjected to the motion compensation prediction in the motion compensation prediction mode. Format conversion 2
7 converts the decoded data to a spatial resolution for processing.

[0098] The video memory 28 temporarily stores the I and P frames for use as reference images in the subsequent decoding processing. The motion compensation prediction 29 performs motion compensation prediction from a prediction mode and a motion vector.

The data decoded in this manner is delayed by the rearrangement buffer 21a and the changeover switch 22a to delay the display of the I frame or the P frame, rearranged in the order of the B frame and rearranged in the correct display order, and output. You.

As described above, the portion from time T1 to time T2 is read from the compressed stream B61 and decoded.

However, the time T1 of the compressed stream B61
In some cases, there is a frame from time T1 to time T2 that depends on a frame other than the part from time T1 to time T2 of the compressed stream. That is, from time T1 to time T2
In some P or B frames up to and including the portion from time T1 to time T2, there may be a frame encoded with reference to an I or P frame.

In such a case, the editing section reading means 9
Expands the range of the portion from time T1 to time T2,
Data of a section including P or I frames other than the portion from time T1 to time T2 on which the P or B frame of the portion from time T1 to time T2 of the compressed stream B61 depends is read. Then, when decoding the part from time T1 to time T2, the decoding means 11 decodes the data of the section whose range has been expanded. Further, of the frames from the time T1 to the time T2, those that depend on the frame before the time T1 are converted into I frames at the time of re-encoding.

The portion of the compressed stream B61 from time T1 to time T2 is passed to the encoding means 14 via the editing means 12.

Here, the editing setting means 8 examines the VBV value immediately before reproducing the data of the 101st frame when the data of the first 100 frames of the compressed stream A60 is reproduced. Then, the VBV immediately before reproducing the data of 101 frames from the head of the compressed stream A60
Assume that the value is Vx. Further, the compressed stream B6
It is assumed that the VBV value obtained when 1 is reproduced from the beginning to time T2 is V2. Then, when re-encoding the portion from time T1 to time T2 of the compressed stream B61, the encoding unit 14 adjusts the encoding amount as follows. That is, the VBV of the compressed stream B61 at the time T1
Assuming that the value is Vx, V
The encoding amount is adjusted so that the BV value becomes V2.

Hereinafter, the VBV value will be described. For this purpose, an operation when the STB 2 reads data and displays an image on a monitor will be described. Thereafter, the encoding unit 14 again determines the V of the compressed stream B61 at the time T1.
Assuming that the BV value is Vx, an operation of adjusting the coding amount so that the VBV value of the portion at time T2 becomes V2 will be described.

Now, as shown in FIG. 5, STB2
The compressed stream is sequentially read from the VCR 4 via the EEE1394 bus 5. The decoding means 50 sequentially inputs the read compressed streams and performs decoding.

That is, as shown in FIG. 6, the demultiplexer 16 of the decoding means 50 separates the transport stream according to the PID attached to the transport packet of the MPEG2 transport stream. The transport buffer 17f has a size of 512 bytes and buffers the bursty high-speed data stream separated by the demultiplexer 16. The multiplex buffer 18f buffers the multiplex jitter of 4 ms and the PES overhead so that the data transfer speed to the next-stage elementary buffer 19f can be consistent. The elementary buffer 19f is
VBV (Video Buffering Verify)
er) It has a buffer function.

The decoder 20f expands the elementary stream. That is, as shown in FIG. 3, the variable length decoding 23 decodes the macroblock coded information, and sets the coding mode, the motion vector, the quantized information, and the quantized DC.
Separate the T coefficients.

The inverse quantization 24 restores DCT coefficients from the decoded 8 × 8 quantized DCT coefficients.

[0110] The inverse DCT 25 performs an inverse DCT transform to convert the data into pixel space data.

The adder 26 outputs the data as it is in the intra-coding mode, and adds and outputs the block data subjected to the motion compensation prediction in the motion compensation prediction mode.

The format converter 27 converts the decoded data into a spatial resolution for processing.

The video memory 28 temporarily stores the I and P frames because they need to be used as reference images in the subsequent decoding processing.

The motion compensation prediction 29 performs the motion compensation prediction from the prediction mode and the motion vector.

Returning to FIG. 6, the rearrangement buffer 21f
Buffer I or P frames. The changeover switch 22f outputs the output of the I frame or the P frame stored in the rearrangement buffer 21 and the decoded B frame.
Switching between frame output and rearrangement in the displayed time order and output.

Returning to FIG. 5, the data thus decoded is converted into an analog signal by the D / A converter 13 and output to the monitor 6. Thus, the monitor 6
Displays an image.

FIG. 7 is a graph in which the amount of accumulation in the elementary buffer 19f of the decoding means 50 of the STB 2 is plotted on the vertical axis and time is plotted on the horizontal axis. Hereinafter, the accumulated amount of the elementary buffer 19f is expressed as VB
It is called V value. As is apparent from FIG. 7, the VBV value changes while taking an intermediate value between the maximum value and zero. When a normal device such as STB2 reproduces a compressed stream, the encoding amount is adjusted during encoding so that the VBV value does not underflow or overflow as shown in the graph of FIG.

Therefore, when the compressed stream edited by the image editing apparatus 1 is reproduced by a normal apparatus such as the STB 2, it is necessary to adjust the encoding amount so that the overflow and underflow of the VBV value do not occur.

The description of the VBV value has been completed.

The encoding means 1 does not overflow or underflow the VBV value.
No. 4 adjusts the coding amount as follows.

That is, as shown in FIG. 4, the format converter 30 of the encoder 14 converts the data into a spatial resolution used for encoding. The screen rearrangement 31 rearranges screens according to I, B, and P frames. The adder 32 calculates a difference between the macroblock video data and the macroblock video data obtained by motion prediction from the reference screen in the case of the motion compensation prediction mode, and outputs the macroblock video data in the case of the intra coding mode.

The DCT 33 performs a DCT transform for each block of 8 pixels × 8 lines, and converts the data into a spatial frequency domain.

The quantization 34 quantizes the converted 8 × 8 DCT coefficients according to target bits and visual characteristics, scans the low-frequency components in order, and converts them into one-dimensional information.

The variable length coding 35 includes macroblock coding information such as a coding mode and a motion vector and quantization DC
The T coefficient is encoded by a variable length code that assigns a shorter code to data having a higher appearance frequency.

The buffer 36 temporarily stores the variable-length coded data.

To perform the above-described processing, the inverse quantization 37 inversely quantizes the quantized DCT coefficients of the I or P frame. Next, the inverse DCT 38 calculates the inversely quantized 8
The inverse DCT transform is performed on the DCT coefficient of × 8 to convert the data into pixel space data. The video memory 40 holds the converted pixel space data as a reference screen. Then, the motion compensation prediction 41 performs motion prediction from the reference screen.

Here, the rate control 43 allocates bits to each frame, performs rate control using a VBV buffer, and performs adaptive quantization using visual characteristics. That is, the code amount is adjusted such that the VBV value at time T2 becomes V2 on the assumption that the VBV value of the data at the time T1 is Vx. First, the quantization 34 adjusts the quantization width at which the quantization is performed. Further, it instructs the DCT coefficient removal 45 to remove high-frequency components from the DCT coefficients. In response to this, the DCT coefficient removal 45 removes high frequency components. Finally, the VBV value at time T2 is V
If it becomes smaller than 2, the rate control 43 instructs the invalid data insertion 44 to insert invalid data. In response to this, invalid data insertion 44 sets V at time T2.
Invalid data is inserted so that the BV value becomes V2.

As described above, the encoding means 14 outputs the VB
Adjust the V value.

The encoding means 14 calculates the re-encoded time T1
From the time T2 to the time T2.
The synthesizing unit 15 combines the data of the compressed stream A60 up to 100 frames with the re-encoded data of the compressed stream B61 from the time T1 to the time T2 whose code amount has been adjusted. Are combined as a compressed stream C62,
Record in CR4.

However, there are cases where up to 100 frames of the compressed stream A60 are dependent on the 101th and subsequent frames. That is, in the B frame of the compressed stream A60, there may be a frame depending on the I or P frame after the 101st frame. In such a case, the editing section reading means 9
The part up to and including the I- or P-frame at the dependency destination after the 101st frame is read. Then, the decoding unit 11 decodes the dependent I or P frame, and decodes the frames dependent on the 101th frame and thereafter by referring to the decoded frame. Further, the encoding unit 14 converts a frame that depends on the 101th frame and subsequent frames into an I-frame and re-encodes it. As described above, the synthesizing unit 15 outputs the compressed stream A6
What is necessary is just to synthesize | combine the compression stream which corrected 0.

When the STB 2 shown in FIG. 5 reads the compressed stream C62 from the VCR 4 and displays it on the monitor 6, the VBV value can be reproduced and displayed normally without underflow or overflow.

In the present embodiment, when two compressed streams are combined, only the section where two I-frames appear is decoded, and the amount of code is adjusted when re-encoding this section. Thus, a compressed stream that can be reproduced and displayed so that the VBV value does not underflow or overflow during reproduction can be generated only by re-encoding only a part of the data.

In this embodiment, the encoding means 14
Has described that, when re-encoding the portion from time T1 to time T2 of the compressed stream B61, motion prediction is performed, and a new motion vector is obtained and encoded. The encoding means 14 determines the time T1 of the compressed stream B61.
When re-encoding the portion from to T2, the motion vector before decoding may be used as it is and re-encoded.
By doing so, the hardware scale can be reduced, and further, image quality degradation can be prevented. Also,
As for the encoding parameters other than the motion vector, those before decoding may be used as they are.

Further, in the present embodiment, two compressed streams are read and one former half and the other latter half are combined to generate one compressed stream. However, the present invention is not limited to this. An equivalent effect can be obtained by reading two compressed streams and editing one after the other.

Further, in the present embodiment, two compressed streams are read from the HDD 4, and a compressed stream generated by combining the first half of one and the second half of the other is a VC stream.
The description has been made on the assumption that the information is recorded in R4, but this is not a limitation. It may be recorded on the HDD 3 again.

Further, in the present embodiment, Vx is converted to VBV at the end when the compressed stream A60 is decoded.
Although described as a value, the present invention is not limited to this. Vx may be a VBV value before or after a predetermined amount from the end when decoding the compressed stream.

(Second Embodiment) Next, a second embodiment will be described.

In the present embodiment, a case where a compressed stream is read from the HDD 3 and a special effect is applied will be described.

As shown in FIG. 1, FIG. 2, FIG. 3, and FIG. 4, the configuration of the image editing apparatus 1 of the present embodiment is the same as that of the first embodiment, so that the description is omitted.

Also, as shown in FIGS. 5 and 6, STB2
Are the same as those in the first embodiment, and a description thereof will be omitted.

Next, the operation of the present embodiment will be described.

In this embodiment, as shown in FIG. 9, when the image editing apparatus 1 reproduces from the beginning of the compressed stream D63, a section reproduced at time T1 and a section reproduced at time T2. Is applied to generate a compressed stream D ″ 65.

For this purpose, the editing and setting means 8 sets the GUI 7
First, the user sets an object to be edited while looking at the GUI screen created in step. That is, a compressed stream to be read is specified. This designation is performed by designating a device to be read and a file name of data stored in the device.

Now, it is assumed that the HDD 3 is designated as the read destination device, and the compressed stream D63 is designated as the file name.

Then, the editing setting means 8 instructs the editing section reading means 9 to read the compressed stream D63 from the HDD 3. The editing section reading means 9 is I
Compressed stream D via EEE1394 I / F10
63 is read from the HDD 3. The read data is passed from the editing setting unit 8 to the decoding unit 11 and is decoded. The GUI 7 creates a GUI screen using the decrypted data. The D / A converter 13 converts this screen into an analog signal, and the monitor 6 displays the analog signal.

The editing setting means 8 sets how to edit the compressed stream D63 with reference to the GUI screen. Now, as shown in FIG.
Suppose that the start of time 3 is set to time 0, and processing for applying a special effect is performed from the portion reproduced at time T1 to the portion reproduced at time T2 when reproduced from the beginning.

Then, the editing setting means 9 instructs to read the portion from the time T1 to the time T2 of the compressed stream D63.

In response to this, the editing section reading means 9
The portion from the time T1 to the time T2 of the compressed stream D63 is read and output to the decoding means 11.

[0149] The decoding means 11 outputs the compressed stream D63.
From time T1 to time T2.
Note that the operation of the decoding unit 11 is the same as that of the first embodiment, and a description thereof will be omitted.

The portion from the time T1 to the time T2 of the compressed stream D63 is sequentially passed to the editing means 12.

The editing means 12 applies a special effect to the portion of the compressed stream D63 from time T1 to time T2. Examples of the special effect include a process of deforming an image or inserting a character. As a result, a portion from time T1 to time T2 of the compressed stream D'64 is obtained.

Here, the editing setting means 8 checks the VBV value of the portion at the time T1 when the portion from the head of the compressed stream D63 to the time T1 is continuously reproduced. Further, the VBV value of the portion at the time T2 when the portion from the head of the compressed stream D63 to the time T2 is continuously reproduced is checked. The VB at the time T1 thus determined
It is assumed that the V value is Vx and the VBV value at time T2 is V2.

However, the compressed bit stream D'64 to which the editing means 12 has applied the special effect is
63 is different from time T1 in time T2.
Therefore, if a simply encoded version of the compressed stream D'64 is reproduced, the VBV value does not become V2 at time T2.

Accordingly, when re-encoding the portion from time T1 to time T2 of the compressed stream D'64, the encoding means 14 adjusts the encoding amount as follows. That is, the VBV value of the compressed stream D'64 at the time T1 is Vx
The encoding amount is adjusted so that the VBV value at the time T2 becomes V2. Encoding means 14
Is the same as that in the first embodiment, and the description is omitted.

The encoding means 14 calculates the time T1
From the time T2 to the time T2.
The synthesizing unit 15 combines the edited and code amount-adjusted data with the portion up to the time T1 of the compressed stream D63,
After that, the data after the time T2 of the compressed stream D63 is combined, and
Record in CR4.

When the STB 2 shown in FIG. 5 reads the compressed stream D ″ 65 from the VCR 4 and displays it on the monitor 6, the VBV value does not underflow or overflow, and can be reproduced and displayed normally.

In this embodiment, after decoding only the section to be edited into one compressed stream, a special effect is applied, and the amount of code is adjusted when the decoded portion is re-encoded.
In this manner, a compressed stream that can be reproduced and displayed so that the VBV value does not underflow or overflow during reproduction can be generated only by re-encoding only the part including the part to be edited. Further, since only a part of the compressed stream is re-encoded, the image quality is less deteriorated.
Further, the compressed data that has been subjected to variable-length encoding can be easily separated at any time, and the compressed data that has been inter-frame encoded can be easily separated at any time.

In this embodiment, after the compressed stream is read from the HDD 3 and subjected to the special effect, the VCR 4
However, the present invention is not limited to this. After reading the compressed stream from HDD 3 and applying special effects,
The information may be recorded on the HDD 3 again.

Further, in the present embodiment, the VBV value at time T1 in the case where the edited and encoded data is decoded as Vx has been described, but the present invention is not limited to this.
VBV before or after a predetermined amount from time T1 when data encoded after editing as Vx is decoded
A value may be used.

(Third Embodiment) Next, a third embodiment will be described.

In the present embodiment, a case where a compressed stream is read from the HDD 3 and a special effect is applied as in the second embodiment will be described.

As shown in FIG. 1, FIG. 2, FIG. 3, and FIG. 4, the configuration of the image editing apparatus 1 of the present embodiment is the same as that of the first embodiment, so that the description is omitted.

As shown in FIGS. 5 and 6, STB2
Are the same as those in the first embodiment, and a description thereof will be omitted.

Next, the operation of the present embodiment will be described.

In the present embodiment, as shown in FIG. 10, when the image editing apparatus 1 reproduces from the beginning of the compressed stream E66, a section reproduced at time T0 and a section reproduced at time T1. Is applied to generate a compressed stream E ″ 68.

For this purpose, the editing setting means 8 sets the GUI 7
First, the user sets an object to be edited while looking at the GUI screen created in step. That is, a compressed stream to be read is specified. This designation is performed by designating a device to be read and a file name of data stored in the device.

Now, it is assumed that the HDD 3 is specified as the read destination device, and the compressed stream E66 is specified as the file name.

Then, the editing setting means 8 instructs the editing section reading means 9 to read the compressed stream E66 from the HDD 3. The editing section reading means 9 is I
Compressed stream E via EEE1394 I / F10
66 is read from the HDD 3. The read data is passed from the editing setting unit 8 to the decoding unit 11 and is decoded. The GUI 7 creates a GUI screen using the decrypted data. The D / A converter 13 converts this screen into an analog signal, and the monitor 6 displays the analog signal.

The edit setting means 8 sets how to edit the compressed stream E66 with reference to the GUI screen. Now, as shown in FIG.
Suppose that the start of time 66 is set to time 0, and the processing for applying the special effect is performed from the part reproduced at time T0 to the part reproduced at time T1 when the reproduction is started from the beginning.

[0170] The editing setting means 9 provides the compressed stream E66.
Is instructed to read from the time T0 part to the time T1 part and from the time T1 part to the time T2 when two I frames have appeared.

In response, the editing section reading means 9
The section from time T0 to time T2 of the compressed stream E66 is read and output to the decoding means 11.

The decoding means 11 outputs the compressed stream E66
From time T0 to time T2.
Note that the operation of the decoding unit 11 is the same as that of the first embodiment, and a description thereof will be omitted.

The portion from time T0 to time T2 of the compressed stream E66 is sequentially passed to the editing means 12.

The editing means 12 applies a special effect to the portion of the compressed stream E66 from time T0 to time T1. Examples of the special effect include a process of deforming an image or inserting a character. As a result, the compressed stream E66 is obtained from the time T0 to the time T0 of the compressed stream E'67 obtained by processing the portion from the time T0 to the time T1.
1 is obtained.

Here, the editing setting means 8 checks the VBV value of the portion at time T1 when the portion from the head of the compressed stream E'67 to time T1 is continuously reproduced. That is, the VBV value at the time T2 is examined when the portion from the time T0 to the time T1 after the special effect is continuously reproduced after the portion before the time T0 is reproduced.
The VBV value at the time T1 checked in this way is Vx
Assume that Further, it is assumed that the VBV value at time T2 is V2 when the compressed stream E66, which is the compressed stream before the special effect is applied, is reproduced from the beginning.

However, the compressed bit stream E'67 to which the editing means 12 has given the special effect is the compressed stream E'67.
66 is different from the time T0 to the time T1.
Therefore, if a simple encoded version of the compressed stream E'67 is reproduced, the VBV value does not become V2 at time T2.

Therefore, when re-encoding the portion of the compressed stream E'67 from time T1 to time T2, the encoding means adjusts the encoding amount as follows. That is, the VBV value of the compressed stream E′67 at the time T1 is Vx
The encoding amount is adjusted so that the VBV value at the time T2 becomes V2. Encoding means 14
Is the same as that in the first embodiment, and the description is omitted.

The encoding means 14 calculates the time T1
From the time T2 to the time T2.
The synthesizing means 15 combines the edited and code amount-adjusted data with the portion of the compressed stream E66 up to time T0,
Thereafter, the data after the time T2 of the compressed stream E66 is combined, and the compressed stream E "68
Record in CR4.

When the STB 2 shown in FIG. 5 reads the compressed stream E ″ 68 from the VCR 4 and displays it on the monitor 6, the VBV value does not underflow or overflow, and can be reproduced and displayed normally.

In this embodiment, after a section to be edited into one compressed stream and a section to adjust the VBV value are decoded, a special effect is applied, and the VBV value is adjusted when the decoded part is re-encoded. The code amount was adjusted in the section. As described above, by simply re-encoding only the part to be edited and the part to adjust the VBV value, it is possible to generate a compressed stream that can be reproduced and displayed so that the VBV value does not underflow or overflow during reproduction.

In this embodiment, after the compressed stream is read from the HDD 3 and subjected to the special effect, the VCR 4
However, the present invention is not limited to this. After reading the compressed stream from HDD 3 and applying special effects,
The information may be recorded on the HDD 3 again.

Further, in the present embodiment, Vx at time T1 when the encoded data after decoding of Vx is decoded is decoded.
Although the description has been made assuming that the BV value is used, the invention is not limited to this. The VBV value before the time T1 by a predetermined amount when decoding the encoded data after editing may be used.

(Fourth Embodiment) Next, a fourth embodiment will be described.

In the present embodiment, a case will be described in which two compressed streams are read from the HDD 3 and transition editing is performed in which one video gradually changes to another video.

As shown in FIG. 1, FIG. 2, FIG. 3, and FIG. 4, the configuration of the image editing apparatus 1 of the present embodiment is the same as that of the first embodiment, so that the description is omitted.

As shown in FIGS. 5 and 6, STB2
Are the same as those in the first embodiment, and a description thereof will be omitted.

Next, the operation of this embodiment will be described.

In the present embodiment, as shown in FIG. 11, when the image editing apparatus 1 reproduces from the beginning of the compressed stream G70, a section reproduced at time T0 and a section reproduced at time T1. Is subjected to transition editing in which the video of the compressed stream F69 gradually changes to the video of the compressed stream G70, and the compressed stream H
The case of generating 71 will be described.

For this purpose, the edit setting means 8 sets the GUI 7
First, the user sets an object to be edited while looking at the GUI screen created in step. That is, a compressed stream to be read is specified. This designation is performed by designating a device to be read and a file name of data stored in the device.

Now, it is assumed that the HDD 3 is specified as a device to be read, and the compressed stream F69 and the compressed stream G70 are specified as file names.

Then, the editing setting means 8 instructs the editing section reading means 9 to read the compressed stream F69 and the compressed stream G70 from the HDD 3. The editing section reading means 9 reads the compressed stream F69 and the compressed stream G70 from the HDD 3 via the IEEE1394 I / F10. The read data is passed from the editing setting unit 8 to the decoding unit 11 and is decoded. GU
I7 creates a GUI screen using the decrypted data. The D / A converter 13 converts this screen into an analog signal, and the monitor 6 displays the analog signal.

The edit setting means 8 sets how to edit the compressed stream F69 and the compressed stream G70 with reference to the GUI screen. Now, as shown in FIG. 11, the start of the compressed stream G70 is set to time 0, and from the part reproduced at time T0 when the compressed stream G70 is reproduced from the start,
The compressed stream F69 is generated before the part reproduced at time T1.
It is assumed that a setting is made so as to generate a compressed stream H71 by performing transition editing in which the video is gradually changed from the video of FIG. The first half of the compressed stream H71 is a compressed stream F69, and the second half is a compressed stream G70.

In FIG. 11, the compressed stream H71
Corresponds to the editing section of the compressed stream G70 and corresponds to the editing section of the compressed stream F69. The VBV matched section of the compressed stream H71 corresponds to the VBV matched section of the compressed stream G70.

[0194] The editing setting means 9 includes a compressed stream G70.
Is instructed to read from the time T0 portion to the time T1 portion and from the time T1 portion to the time T2 when two I frames have appeared. Further, from the time T0 portion of the compressed stream G70 to the time T0
It instructs to read the edit section of the compressed stream F69 to be combined with the portion 1.

In response, the editing section reading means 9
The portion from the time T0 to the time T2 of the compressed stream G70 is read and output to the decoding means 11. In addition, the editing section of the compressed stream F69 combined with the compressed stream G70 is read and output to the decoding means 11.

The decoding means 11 outputs the compressed stream G70
From time T0 to time T2.
Further, the editing section of the compressed stream F69 is decoded. Note that the operation of the decoding unit 11 is the same as that of the first embodiment, and a description thereof will be omitted.

However, similar to the first embodiment, there is a case where a portion of the compressed stream G70 from time T0 to time T2 depends on a frame other than the portion from time T0 to time T2 of the compressed stream. is there. That is, there is a case where a P or B frame in a portion from time T0 to time T2 is coded with reference to an I or P frame other than the portion from time T0 to time T2.

In such a case, the editing section reading means 9
Expands the range of the portion from time T0 to time T2,
The data of the section including the P or I frame other than the portion from the time T0 to the time T2 on which the P or B frame of the portion from the time T0 to the time T2 of the compressed stream G70 depends is read. Then, when decoding the part from time T0 to time T2, the decoding means 11 decodes the data of the section whose range has been expanded. Further, of the frames from the time T0 to the time T2, those that depend on the frames before the time T0 are converted into I frames at the time of re-encoding.

The portion of the compressed stream G70 from time T0 to time T2 and the portion of the compressed stream F69 in the editing section are sequentially passed to the editing means 12.

The editing means 12 performs transition editing on the portion of the compressed stream G70 from time T0 to time T1. That is, at time T0, the compressed stream F6
9 is displayed, and the video of the compressed stream G70 is displayed with weight as it approaches time T1. Then, at time T1, synthesis is performed so that the video of the compressed stream G70 is displayed. As a result, the compressed stream G
At 70, processing is applied to a portion from time T0 to time T1, and an edited section of the compressed stream H71 is obtained. The compressed stream H71 is a compressed stream after the transition editing. From the beginning to the edited section, the compressed stream F69 is a compressed stream F69.
This is a part that gradually changes to 70, and the part after the edited section is the compressed stream G70.

Here, the editing setting means 8 sets the VB of the last part of the edited section when the part from the beginning of the compressed stream H71 to the end of the edited section is continuously reproduced.
Check the V value. It is assumed that the VBV value of the last portion of the edited section checked in this way is Vx. Further, it is assumed that the VBV value at the end of the VBV-matched section when the compressed stream G70, which is the compressed stream before the special effect is applied, is V2.

However, the compressed bit stream H71 to which the editing means 12 has given the special effect is the compressed stream G7.
The part up to the end of the edited section is different from 0.
Therefore, if a simple encoded version of the compressed stream H71 is played back, the VBV-matched section ends with a VBV matched section.
The V value does not become V2.

Accordingly, when re-encoding the VBV-matched section of the compressed stream H71, the encoding unit 14 adjusts the encoding amount as follows. That is, assuming that the first VBV value of the VBV matched section is Vx,
The coding amount is adjusted so that the VBV value of the VBV-matched section becomes V2. The method by which the encoding unit 14 adjusts the encoding amount is the same as that in the first embodiment, and a description thereof will be omitted. At the same time as performing such an encoding amount, the edited section is also encoded.

The encoding means 14 outputs the re-encoded data of the VBV matched section to the synthesizing means 15. The combining means 15 combines the first half of the compressed stream F69 before the edited section of the compressed stream H71, combines the second half of the compressed stream G70 after the VBV matching section, and records the compressed stream H71 on the VCR4. .

[0205] However, there may be a frame up to the end of the editing section of the compressed stream F69, which depends on a frame after the editing section. That is, in the B frame of the compressed stream F69, there may be a frame depending on the I or P frame after the editing section. In such a case, the editing section reading means 9 reads up to the portion including the dependent I or P frame after the editing section. Then, the decoding unit 11 decodes the dependent I or P frame and refers to the decoded frame to decode the frames that depend on the edit section and thereafter. Further, the encoding unit 14 converts a frame depending on a frame after the editing section into an I frame and re-encodes the frame. in this way,
The synthesizing unit 15 may synthesize a compressed stream obtained by modifying the compressed stream F69.

When the STB 2 shown in FIG. 5 reads the compressed stream H71 from the VCR 4 and displays it on the monitor 6, the VBV value does not underflow or overflow, and can be reproduced and displayed normally.

In the present embodiment, when transition editing is performed on two compressed streams, and a transition editing section and a VBV value adjusting section are decoded, transition editing is performed and the decoded part is re-encoded. In the section where the VBV value is adjusted, the code amount was adjusted. As described above, by simply re-encoding only the part to be edited and the part to adjust the VBV value, it is possible to generate a compressed stream that can be reproduced and displayed so that the VBV value does not underflow or overflow during reproduction.

In this embodiment, after the compressed stream is read from the HDD 3 and subjected to the special effect, the VCR 4
However, the present invention is not limited to this. After reading the compressed stream from HDD 3 and applying special effects,
The information may be recorded on the HDD 3 again.

(Fifth Embodiment) Next, a fifth embodiment will be described.

In this embodiment, a case where a compressed stream is read from the HDD 3 and a telop is synthesized will be described.

FIGS. 12, 2, 3 and 4 show an image editing apparatus 52 according to this embodiment. The difference from the first embodiment is that the image editing device 52 of the present embodiment includes a changeover switch 51.

The changeover switch 51 detects a macroblock including a telop from the compressed stream read by the editing section reading means 9 and decodes the macroblock and a macroblock which refers to the macroblock as a motion vector. 11 is a means for sending a macroblock containing no telop to the editing means 12 directly without passing through the decoding means 11.

As shown in FIGS. 5 and 6, STB2
Are the same as those in the first embodiment, and a description thereof will be omitted.

Next, the operation of the present embodiment will be described.

In this embodiment, the image editing device 52
As shown in FIG. 13A, a telop is inserted into a section between a part reproduced at time T1 and a part reproduced at time T2 when reproduced from the beginning of the compressed stream I72,
A case where the compressed stream I'73 is generated will be described.

For this purpose, the editing setting means 8 sets the GUI 7
First, the user sets an object to be edited while looking at the GUI screen created in step. That is, a compressed stream to be read is specified. This designation is performed by designating a device to be read and a file name of data stored in the device.

Now, it is assumed that the HDD 3 is specified as the device to be read, and the compressed stream I72 is specified as the file name.

Then, the editing setting means 8 instructs the editing section reading means 9 to read the compressed stream I72 from the HDD 3. The editing section reading means 9 is I
Compressed stream I via EEE1394 I / F10
72 is read from the HDD 3. The read data is passed from the editing setting unit 8 to the decoding unit 11 and is decoded. The GUI 7 creates a GUI screen using the decrypted data. The D / A converter 13 converts this screen into an analog signal, and the monitor 6 displays the analog signal.

The editing setting means 8 sets how to edit the compressed stream D63 with reference to the GUI screen. Now, as shown in FIG.
Suppose that the start of 72 is set to time 0, and the processing for inserting the telop from the part reproduced at time T1 to the part reproduced at time T2 when the reproduction is started from the beginning is set.

[0220] The editing setting means 9 includes a compressed stream D63.
From the time T1 to the time T2.

In response to this, the editing section reading means 9
The portion from the time T1 to the time T2 of the compressed stream D63 is read and output to the changeover switch 51. The changeover switch 51, as shown in FIG.
4 is detected, and the macroblock and the macroblock that refers to the macroblock as a motion vector are output to the decoding unit 11.
Further, as shown in FIG. 13B, a macroblock 76 which does not include the telop 74 is detected, and the macroblock is directly passed to the editing unit 12 without passing through the decoding unit 11.
Output to

[0222] The decoding means 11 generates the compressed stream 72
The macroblock inputted from time T1 to time T2 is decoded. Note that the decryption means 11
Are the same as those in the first embodiment, and a description thereof will be omitted.

The macroblock input to the decoding means 11 out of the portion from time T1 to time T2 of the compressed stream I72 is decoded and passed to the editing means 12.

[0224] The editing means 12 inserts a telop in the portion of the compressed stream I72 from time T1 to time T2. Since a macroblock including a telop has been decoded, a telop is inserted for each decoded macroblock. As a result, the time T1 of the compressed stream I'73
From time T2 to time T2.

Here, the editing setting means 8 checks the VBV value of the portion at the time T1 when the portion from the head of the compressed stream I63 to the time T1 is continuously reproduced. The VBV value of the portion at the time T2 when the portion from the beginning of the compressed stream 72 to the time T2 is continuously reproduced is checked.
The VBV value at the time T1 checked in this way is Vx
And the VBV value at time T2 is V2.

[0226] However, the compressed bit stream I'73 to which the editing means 12 has applied the special effect is different from the compressed stream 72 in that a telop is inserted from time T1 to time T2. Therefore, if a simply encoded version of the compressed stream I'73 is reproduced, the VBV value does not become V2 at time T2.

Therefore, the encoding means 14 adjusts the encoding amount as follows when re-encoding the macroblock in which the telop is inserted in the portion from time T1 to time T2 of the compressed stream I'73. . That is, assuming that the VBV value of the compressed stream I'73 at the time T1 is Vx, the encoding amount is adjusted so that the VBV value of the portion at the time T2 becomes V2. The method by which the coding unit 14 adjusts the coding amount is the same as that of the first embodiment except that the coding amount is adjusted in units of macroblocks, and therefore the description is omitted.

The encoding means 14 calculates the time T1
From the time T2 to the time T2.
The synthesizing unit 15 combines the edited and code amount-adjusted data into the portion of the compressed stream I72 up to time T1,
After that, the data after the time T2 of the compressed stream I72 is combined, and the compressed stream
Record in CR4.

When the STB 2 shown in FIG. 5 reads the compressed stream I'73 from the VCR 4 and displays it on the monitor 6, the VBV value does not underflow or overflow, and can be reproduced and displayed normally.

In the present embodiment, after decoding only a macroblock including a telop in a section to be edited into one compressed stream and a macroblock referring to the macroblock as a motion vector, the telop is inserted and the decoded macroblock The code amount was adjusted when re-coding. In this way, a compressed stream that can be reproduced and displayed so that the VBV value does not underflow or overflow during reproduction can be generated only by re-encoding only the macroblock including the telop.

The IEEE 1394I of the present embodiment
/ F10 and the editing section reading means 9 are examples of the input means of the present invention, and the decoding means 11, editing means 12, encoding means 14, and synthesizing means 15 of the present embodiment are examples of the editing and adjusting means of the present invention. The synthesizing unit 15, the editing section reading unit 9, and the IEEE 1394 I / F 10 of the present embodiment are examples of the output unit of the present invention. The decoding unit 11, the editing unit 12, and the encoding unit 14 of the present embodiment are examples of the editing unit of the present invention.

Further, in this embodiment, after the compressed stream is read from the HDD 3 and subjected to the special effect,
4, but it is not limited to this. After reading the compressed stream from the HDD 3 and applying the special effect, the compressed stream may be recorded on the HDD 3 again.

Further, in the present embodiment, V at time T2
Although the BV value has been described as V2, the present invention is not limited to this.
A value may be used.

Further, the image editing may be performed a plurality of times by combining the image editing performed in each of the first to fifth embodiments.

Further, in the present embodiment, the description has been made assuming that the compressed stream has the format of the MPEG2 transport stream, but the present invention is not limited to this. What is necessary is just to use a format that compresses image data, such as an MPEG2 program stream or an MPEG1 format stream.

Furthermore, the compressed stream of the present embodiment is an example of the compressed bit stream of the present invention.

Further, all or some of the functions of all or part of the image editing apparatus of the present invention may be realized by hardware, or may be realized by software using a computer program.

Further, all or some of the operations of all or some of the steps of the image editing method of the present invention may be realized by hardware, or may be realized by software using a computer program.

Further, the present invention is a program recording medium for recording a program and / or data for causing a computer to execute all or a part of the functions of all or a part of the image editing apparatus of the present invention. The present invention also includes a program recording medium, wherein the program and / or data read out execute the function in cooperation with the computer.

Further, the present invention is a program recording medium for recording a program and / or data for causing a computer to execute all or a part of the operations of all or some of the steps of the image editing method of the present invention. The present invention also includes a program recording medium, wherein the program and / or data read out execute the function in cooperation with the computer.

[0241]

As is apparent from the above description, the present invention provides an image editing method, an image editing apparatus, and a program recording medium with little deterioration in image quality even when compressed data is edited. I can do it.

Further, the present invention can provide an image editing method, an image editing apparatus, and a program recording medium capable of easily separating compressed data that has been subjected to variable length encoding at an arbitrary time.

Further, the present invention can provide an image editing method, an image editing apparatus, and a program recording medium capable of easily separating compressed data that has been encoded between frames at an arbitrary time.

Further, the present invention can provide an image editing method, an image editing apparatus, and a program recording medium which do not break the buffer of the decoder even if any compressed data is cut.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of an image editing apparatus according to first to fourth embodiments of the present invention.

FIG. 2 is a block diagram showing a configuration of a decoding unit according to the first to fifth embodiments of the present invention;

FIG. 3 is a block diagram showing a configuration of a decoder according to the first to fifth embodiments of the present invention.

FIG. 4 is a block diagram showing a configuration of an encoding unit in the first to fifth embodiments of the present invention.

FIG. 5 is an STB according to the first to fifth embodiments of the present invention.
Block diagram showing the configuration of

FIG. 6 is a block diagram showing a configuration of a decoding unit according to the first to fifth embodiments of the present invention;

FIG. 7 shows a VBV according to the first to fifth embodiments of the present invention.
Diagram showing time change of values

FIG. 8 is a view for explaining editing of a compressed stream in the first embodiment of the present invention.

FIG. 9 is a view for explaining editing of a compressed stream in the second embodiment of the present invention.

FIG. 10 is a diagram illustrating editing of a compressed stream according to the third embodiment of the present invention.

FIG. 11 is a diagram illustrating editing of a compressed stream according to the fourth embodiment of the present invention.

FIG. 12 is a block diagram illustrating a configuration of an image editing device according to a fifth embodiment of the present invention.

FIG. 13 is a view for explaining editing of a compressed stream in the fifth embodiment of the present invention.

[Explanation of symbols]

 Reference Signs List 1 image editing apparatus 2 STB 3 HDD 4 VCR 5 IEEE1394 bus 6 monitor 7 GUI 8 editing setting means 9 editing section reading means 10 IEEE1394 I / F 11 decoding means 12 editing means 13 D / A conversion 14 encoding means 15 combining means 17a To f transport buffer 18a to f multiplex buffer 19a to f elementary buffer 20a to f decoder 21a to f rearrangement buffer 22a to f switch 23 variable length decoding 24 inverse quantization 25 inverse DCT 26 adder 27 format conversion 28 Video memory 29 Motion compensation prediction 30 Format conversion 31 Screen rearrangement 32 Adder 33 DCT 34 Quantization 35 Variable length coding 36 Buffer 37 Inverse quantization 38 Inverse DCT 39 Adder 40 Video memory 41 Motion compensation prediction Measurement 42 Changeover switch 43 Rate control 44 Invalid data insertion 45 DCT coefficient removal

Continuing on the front page (72) Inventor Kazuhiro Waki 1006 Kazuma Kadoma, Osaka Prefecture Matsushita Electric Industrial Co., Ltd. (Reference) 5C052 AA01 GA03 GB06 GD03 GE04 5C053 FA14 FA23 GA14 GB08 GB17 GB19 GB21 GB22 GB26 GB38 KA03 LA06 LA07 5C059 KK35 KK39 LA01 MA00 MA04 MA05 MA23 ME01 PP05 PP06 PP07 SS20 UA02 UA05 UA33

Claims (21)

[Claims] 1. An input means for reading a compressed bit stream, decoding a section from time T1 to T2 in the input compressed bit stream, and performing or editing the decoded image for a first time. When the compressed bit stream obtained by performing the encoding is decoded again when encoding is performed at the time T2,
o Editing adjusting means for adjusting the code amount so that the value of the Buffering Verifyer (hereinafter referred to as VBV value) becomes a predetermined V2; and output means for outputting the compressed bit stream output by the editing adjusting means. It is assumed that the VBV value at the time T1 is Vx, and the V2 is the VBV at the time T2 when the input compressed bit stream is decoded as it is.
An image editing apparatus characterized by being a value. 2. The image editing apparatus according to claim 1, wherein the first editing is performed over all or only a part of the time from the time T1 to the time T2. 3. The Vx value is the VBV value at the end of another compressed bit stream to be combined with the compressed bit stream, that is, the VBV value at the end when the another compressed bit stream is decoded as it is.
2. The image editing apparatus according to claim 1, wherein the value is a V value. 4. When the edit adjustment means performs the first edit between the time T1 and the time T2, the Vx is a VBV value of the input compressed bit stream at the time T1. 2. The VBV value when the input compressed bit stream is decoded until the time T1.
An image editing apparatus as described in the above. 5. The edit adjusting means decodes the input compressed bit stream for a predetermined section between the time T0 and the time T1 before the time T1, and converts the decoded data into decoded data. Vx is a VBV at the time T1 when the compressed bit stream decoded after the second editing is decoded.
2. The image editing apparatus according to claim 1, wherein the value is a value. 6. The second editing is a transition editing in which the video of the other compressed bit stream is gradually switched to the video of the input compressed bit stream, and the data before the time T0 is data of the input. 6. The image according to claim 5, wherein a portion of the compressed bit stream before the time T0 is not used, and a portion of the other compressed bit stream before the portion used for the transition editing is used instead. Editing device. 7. The image according to claim 2, wherein the first editing is performed in a part of the time from the T1 time to the T2 time, and the decoding is performed on a part related to the editing. Editing device. 8. The first editing is to insert a telop, and the part to be decoded is a macroblock unit. Not only the macroblock into which the telop is inserted but also the macroblock as a motion vector. 8. The image editing apparatus according to claim 7, wherein a macro block to be referred to is also decoded. 9. The section for adjusting the VBV value, when performing re-encoding, performs re-encoding using an encoding parameter before decoding as it is. Image editing device. 10. The image editing apparatus according to claim 1, wherein the code amount is adjusted by adjusting a quantization level. 11. The method according to claim 1, wherein the code amount is adjusted by selecting a DCT coefficient.
0. The image editing device according to any one of 0. 12. The image editing apparatus according to claim 1, wherein the adjustment of the code amount is performed using dummy data. 13. The VBV value at a predetermined time before or after the T2 time may be used as the V2 value instead of the VBV value at the T2 time.
An image editing device according to any one of the above. 14. The method according to claim 1, wherein the Vx is a VBV value before or after a predetermined amount from the end, instead of the VBV value at the end when the another compressed bit stream is decoded. The image editing apparatus according to claim 3, wherein 15. The VxV is a VBV value of the input compressed bit stream at the time T1, instead of the VBV value of the input compressed bit stream, a predetermined amount before or after the time T1 of the input compressed bit stream. 5. The image editing apparatus according to claim 4, wherein the VBV value of the image editing apparatus may be used. 16. The Vx is VB at the T1 time when the second post-encoded data is decoded.
6. The image editing device according to claim 5, wherein the VBV value may be a VBV value that is a predetermined amount before the time T1 when the second post-encoded data is decoded, instead of the V value. . 17. An input means for reading a compressed bit stream, and when combining the read first compressed bit stream and the second compressed bit stream, a first half of the first compressed bit stream and the second When merging the second half of the compressed bit stream, the first half of the first compressed bit stream
If there is a frame that depends on the second half of the compressed bit stream, the second frame of the first compressed bit stream is used to decode and re-encode the frame that depends on the second half. The second part of the compressed bit stream
When there is a frame that depends on the first half of the compressed bit stream, editing means that decodes and re-encodes the frame that depends on the first half using the first frame of the second compressed bit stream; Output means for outputting the combined compressed bit stream. 18. Decoding a section from time T1 to T2 in the predetermined compressed bit stream that is input,
When the decoded image is encoded with or without first editing, and the compressed bit stream obtained by performing the encoding is decoded again, the VBV value at the time T2 is used. Is an image editing method for adjusting the code amount such that the VBV value at the time T1 is Vx. The V2 is the input compressed bit stream as it is. VBV at the time T2 when decrypted
An image editing method characterized by being a value. 19. When merging a first compressed bit stream and a second compressed bit stream, when merging a first half of the first compressed bit stream and a second half of the second compressed bit stream, The first part of the first compressed bit stream includes the first
If there is a frame that depends on the second half of the compressed bit stream, the second frame of the first compressed bit stream is used to decode and re-encode the frame that depends on the second half. The second part of the compressed bit stream
When there is a frame that depends on the first half of the compressed bit stream, the first frame of the second compressed bit stream is used to decode and re-encode the frame that depends on the first half. How to edit images. 20. Program recording for recording a program and / or data for causing a computer to execute all or a part of the functions of all or a part of the image editing apparatus according to any one of claims 1 to 17. A program recording medium, which is readable by a computer. 21. A program recording recording a program and / or data for causing a computer to execute all or part of all or part of the steps of the image editing method according to claim 18 or 19. A program recording medium, which is readable by a computer.