JP2000278643A - Data conversion device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To generate intermittent video sound without decoding and encoding again encoded video information by separating first video information and encoded non-video information and converting encoded first video information into encoded second video information. SOLUTION: A separation part 22 receives MPEG video sound information from an encoded video sound input part 20 and separates it into encoded video information and encoded non-video information. The separation processing is executed in the form of digital data without decoding it. Separated video information is outputted to a video conversion part 24 and non-video information is outputted to a connection part 26. The separation part 22 outputs information showing by what kind of relation video information and non-video information constitute MPEG video sound information to the video conversion part 24 as connection information. The video conversion part 24 receives encoded video information from the separation part 22 and converts received video information into different video information.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for recording and reproducing digitally compressed and encoded video and audio data intermittently for video and audio continuously.

[0002]

2. Description of the Related Art In recent years, there has been an increasing number of occasions in which images such as satellite digital broadcasts and DVDs are distributed in the form of digital data. In the future, terrestrial digital broadcasting is also planned, and it is easily expected that digitalization of video will be further spurred. Therefore, it is considered that digitized video information will be used in many video-related fields in the future.

[0003] Processing for handling digitized video information, for example, processing with a relatively large load such as non-linear video editing and video search, is processing that can be performed only by professionals in the video industry having expensive editing equipment. there were. However, in recent years, with the spread of high-performance personal computers and the like capable of processing large amounts of data at high speed, the amount of information that can be handled by ordinary individuals has increased dramatically. As a result, the processing with a large load described above is becoming a category of processing that can be performed by an individual at home. It will only be a matter of time before the lifestyle in which ordinary individuals receive, store, edit and use a large amount of video information including audio information is established.

[0004] However, personal time that can be spent on information collection is often limited, and the amount of information that can be processed within that time is limited, so that a large amount of transmitted information cannot be used effectively. There is a fear. Therefore,
There is a need for technology that enables efficient use of only information important to individuals.

In using only important information, a problem in processing video information or audio information in a portable terminal is a data size of the video information or audio information. Storing video information or audio information in a portable terminal as it is is often too much and wasteful in terms of capacity.

Therefore, it would be useful to automatically sort information according to its importance so that it can be referenced quickly anywhere. In particular, in recent years, with the spread of mobile devices, information can be easily referred to even on a mobile terminal. For example, by storing video or audio information on a personal computer at home, searching for important video or audio information, and then copying it to a mobile terminal, you can easily refer to the information when you are away from home, etc. can do.

In order to make the capacity of video information enough to be stored even in a portable terminal, a continuous video can be converted into an intermittent video such as a frame-by-frame video, which is a thinned-out continuous video. . Even in the case of intermittent video, the user can sufficiently utilize information. This is because the information required by the user is often sufficient for a representative video frame in the continuous video information, and need not always be provided as a continuous video. Therefore, an efficient thinning processing method for thinning a large amount of video information is desired.

Japanese Patent Laid-Open Publication No. Hei 8-98133 discloses a system that efficiently uses a storage area by recording video intermittently and audio continuously.

FIG. 9 shows a block diagram of a system disclosed in Japanese Patent Application Laid-Open No. Hei 8-98133.

The video data 101 is stored in the video data recording unit 1
02 and the scene change detection unit 103. The audio data 104 is input to the scene change detection unit 103 and the audio data recording unit 105.

The scene change detecting section 103 detects a scene change using the image memory 121 and the data comparing section 122. The scene change is detected by the video data 1
01 and the audio data 104. The video data 101 at the time of detecting a scene change is data representing a video itself. That is, the video data 101 is not video data that has been digitally encoded but is video data that has been decoded. The data comparison unit 122 of the scene change detection unit 103 outputs a detection signal 106 indicating that a scene change has been detected to the recording control unit 107.

The recording control unit 107 outputs a video data recording signal 108 and reproduction timing information 109 based on the detection signal 106. The video data recording signal 108 is a signal for controlling the video data recording unit 102. The recording control unit 107 inputs the reproduction timing information 109 to the control information recording unit 112. The control information recording unit 112
After performing signal processing necessary for recording the input reproduction timing information 109 in the semiconductor memory 111,
The reproduction timing information 109 is written in the semiconductor memory 111.

The video data recording section 102 receives the video data 101 from the input terminal or the video data 1 from the scene change detection section 103 related to the video data recording signal 108.
The signal processing required for recording on the semiconductor memory 111 is performed on the data 10. The video data recording unit 102 writes the signal-processed video data 101 or video data 110 in a semiconductor memory 111 that is a recording medium.

The audio data 104 is stored in the audio data recording unit 1
At 05, the signal processing required for recording is performed and written into the semiconductor memory 111.

The above-mentioned Japanese Patent Laid-Open Publication No. Hei 8-98133 is a technique in which audio is continuously recorded, while video is recorded only when there is a change. Since there is no information in the intermittent part of the intermittently recorded video, there is a problem that the semiconductor memory is wasted. In the above-mentioned system, in order to eliminate this waste, a portion where no information exists is omitted, and a gap between video data is recorded in a narrowed state. In order to close the gap between the video data and record, it is necessary to synchronize the video and the audio. Therefore, when the intermittent video and the audio are recorded, the synchronization information of the video and the audio is also recorded.

[0016]

However, since the intermittent video / audio information output by the above system has its own format, it restricts the user's use environment and is inconvenient.

Further, in the above system, digitized video information must first be decoded in order to detect a scene change. Since this decoding is necessary for all digital information, the amount of processing is large.

In addition, the decoding process causes image quality degradation when writing the video used for detecting a scene change to the storage device.

Further, in the final stage of the processing, it is necessary to re-encode the video and the audio, and the processing amount is large.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and uses the data structure of video and audio information encoded according to the MPEG standard to decode and decode the encoded video and audio information. MP without re-encoding
An object is to create intermittent video and audio that conforms to the EG standard.

[0021]

According to the present invention, there is provided a data conversion apparatus comprising: first data which includes encoded first video information representing video and encoded non-video information representing information other than video; An input unit for inputting a sequence, and without decoding the encoded first video information and the encoded non-video information, the encoded first video information and the encoded first video information included in the first data sequence. 1 video information and the coded non-video information, and a video conversion unit that converts the separated coded first video information into coded second video information. And combining the encoded second video information with the separated encoded non-video information,
A data conversion device having a combining unit that generates a second data string, whereby the above object can be achieved.

[0022] The encoded second video information may represent an intermittent video obtained by thinning the video.

The separating section outputs first combined information relating to combining the separated coded first video information and the coded non-video information to the video converting section;
The video converter is configured to convert the encoded second video information from the first combined information in accordance with the conversion from the encoded first video information to the encoded second video information. Generating second combination information relating to a combination of information and the encoded non-video information, wherein the coupling unit generates the second video information encoded based on the second combination information and the code The second data string may be generated by combining the converted non-video information.

[0024] The encoded non-video information includes the first
Including synchronization information for synchronizing the reproduction of the encoded first video information and the reproduction of the encoded non-video information included in the data sequence, and wherein the combining unit converts the synchronization information into It may be used as synchronization information of the encoded second data.

The first data stream is a first MPEG data stream that has been MPEG-compressed and encoded, and the second data stream is a second MPEG data stream that has been MPEG-compressed and encoded. The encoded first video information may include a predetermined number of first pictures, and the encoded second video information may include the predetermined number of second pictures.

[0026] The video converter converts the at least one picture and the dummy picture out of the predetermined number of the first pictures excluding at least one picture into a dummy picture. Generating a predetermined number of the second pictures, wherein the at least one picture is a picture including information representing a part of the video, and the dummy picture is information for holding a display of a part of the video. May be included.

[0027] The predetermined number of first pictures includes at least an intra picture and a non-intra picture,
The intra picture and the non-intra picture may be arranged in a predetermined order, and the at least one picture may include an intra picture arranged first.

The first data sequence is a first MPEG data sequence that has been MPEG-compressed and coded, and the second data sequence is a second MPEG data sequence that has been MPEG-compressed and coded. The first video information includes a plurality of GOPs each having a plurality of pictures, and the video conversion unit includes a picture excluding at least one picture of a first GOP among the plurality of GOPs, GOP
Generating a single GOP from at least one picture of the first GOP and the dummy picture by converting the plurality of pictures included in each of the plurality of GOPs to a dummy picture, 1 GO
The at least one picture of P is a picture including information representing a part of the video, the dummy picture is a picture including information for holding a display of a part of the video, and the plurality of GOPs The total number of the plurality of pictures included may be the same as the total number of pictures included in the single GOP generated by the video conversion unit.

[0029] Each of the plurality of GOPs is arranged in a first predetermined order, and the plurality of pictures included in each of the plurality of GOPs include at least an intra picture and a non-intra picture; The picture and the non-intra picture are arranged in a second predetermined order, and the first GOP is a first GOP among the plurality of GOPs arranged in the first predetermined order. And at least one picture of the first GOP may be an intra picture arranged first among the plurality of pictures of the first GOP arranged in the second predetermined order. .

[0030] The encoded first video information is arranged in a predetermined order, and includes a plurality of GOPs each including at least an intra picture. A partial decoding unit that decodes a predetermined portion of information, and further includes, based on the decoded predetermined portion, an event detection unit that detects an event representing an image in which characteristics of the video greatly change, The video converter may select at least one picture based on a picture corresponding to the event and generate the predetermined number of the second pictures.

[0031] The at least one picture may be an intra picture of a GOP arranged next to a GOP including a picture corresponding to the event.

[0032] The at least one picture may be a picture next to a picture corresponding to the event.

[0033] The at least one picture may be a picture that is played after a predetermined time has elapsed since the picture corresponding to the event was played.

[0034] The data conversion device may use an arbitrary picture among the pictures arranged in the predetermined order as the at least one picture according to a predetermined parameter.

The coded non-video information includes the first
The first packet length data indicating the information amount of the second video information, and the combining unit modifies the first packet length data in accordance with the information amount of the second video information to thereby obtain a second packet length. Long data may be generated and included in the second data string as the non-video information.

The operation will be described below.

According to the invention of claim 1 of the present invention, the encoded first video information and the encoded non-video information are not decoded and the code included in the first data string is not decoded. The encoded first video information is separated from the encoded non-video information, and the encoded first video information is converted into the encoded second video information. This makes it possible to obtain video information without image quality degradation due to decoding. Furthermore, the load required for the calculation required for decoding can be reduced.

According to the invention according to claim 3 of the present invention, the video conversion unit performs the first combination according to the conversion from the encoded first video information to the encoded second video information. From the information, second combined information relating to a combination of the encoded second video information and the encoded non-video information is generated. By using the second combined information, it is possible to combine the second video information and the non-video information even if the encoded first video information is converted to the encoded second video information. it can.

According to the invention of claim 4 of the present invention, the synchronization information for synchronizing the reproduction of the encoded first video information and the reproduction of the encoded non-video information is encoded. This is used as synchronization information for the second data. Thereby, even if the first video information is converted to the second video information, the same synchronization information can be used, and it is not necessary to perform a process of newly generating the synchronization information.

According to the fifth aspect of the present invention, M
The first video information and the second video information of the PEG compression-encoded data string include the same number of pictures. Thus, even if the first video information is converted to the second video information, video information having the same reproduction time can be obtained.

According to the eighth aspect of the present invention, the video converter converts the first video information including a plurality of GOPs into the second video information including a single GOP. Thus, even if the first video information is converted to the second video information, video information having the same reproduction time can be obtained.

According to the fifteenth aspect of the present invention,
The combining unit corrects the first packet length data indicating the information amount of the first video information according to the information amount of the second video information, and generates second packet length data. The generated second packet length data is included in the second data string as non-video information. As a result, the first video information is
, The second data string generated based on the second video information can be appropriately reproduced.

[0043]

DESCRIPTION OF THE PREFERRED EMBODIMENTS In a data converter according to the present invention, a data string containing video information and / or audio information to be inputted is a digital data string having a data structure of the MPEG standard, for example. In the following, description will be made assuming that the input data string is a digital data string encoded according to the MPEG standard (hereinafter, referred to as “MPEG data string”). In the MPEG data sequence, particularly, the MPEG-encoded video / audio information is referred to as MPEG video / audio information. MP
The EG video and audio information is distributed by media such as satellite broadcasting and DVD.

Needless to say, if the method described below can be applied, compressed data in another digital format other than the MPEG data string may be used.

First, the structure of an MPEG data string handled by the data converter according to the present invention will be described.

FIG. 1 shows a general data structure (data format) of an MPEG data string.

The MPEG data string includes a plurality of packs 11 which are data blocks. The pack 11 includes a pack header 12, a system header 13, and one or more packets 14.

The pack header 12 includes buffer synchronization information (not shown) for synchronizing buffers. The system header 13 includes system information (not shown) of each pack.
including. The packet 14 includes a packet start code 15 called a packet header, a packet length 16 indicating the data length of the packet, synchronization information 17, and packet data 18.

The packet start code 15 is a data block having a 32-bit width. The packet start code 15 is information indicating the start of a packet, and by looking at the lower 8 bits, it is possible to identify whether the packet is a video packet or an audio packet. The synchronization information 17 is used for synchronizing the reproduction of the data included in the packet with the data of another packet during reproduction. The packet data 18 includes attribute information indicating the attribute of the packet data 18 in addition to the information indicating the actual video or audio.

Hereinafter, the video information is the packet 1 identified as a video packet by the packet start code 15.
4 refers to information included only in the packet data 18.

The voice information is information included only in the packet data 18 of the packet 14 identified as a voice packet by the packet start code 15.

The non-video information is information other than the video information in the MPEG video / audio information, that is, the pack header 12, system header 13,
Packet start code 15, packet length 16, synchronization information 1
7 and information included in the packet data 18 of the packet identified as audio information.

(Embodiment 1) Hereinafter, a data conversion apparatus according to a first embodiment of the present invention will be described with reference to the drawings.

1. Overview of Operation of Data Conversion Apparatus of the Present Invention The data conversion apparatus of the present invention provides a GOP (Group Of Pic) of input MPEG video information representing continuous video.
This is a device that converts the data into intermittent video information while maintaining the structure. Here, the GOP refers to a unit of data in which a plurality of coded pictures (one screen or one frame of video) of several frames are collected.

A GOP referred to in the following description is a GOP.
It includes a header, an I picture, a B picture, and a P picture (a specific data structure of a GOP is illustrated in FIG. 5B). The GOP header records GOP attributes, time codes, and the like. An I picture represents an intra-frame coded image, a B picture represents a bidirectional predictive coded image, and a P picture represents an inter-frame forward predicted coded image. Note that I
The picture is an intra picture, and the P and B pictures are non-intra pictures.

The data conversion device includes a GOP header and a GOP
Are replaced with dummy P pictures except for the first I picture. The dummy P picture is a picture including only information indicating that there is no change from the previous frame, that is, only information for holding display of a frame representing a part of a video. Since the video information represented by the I picture is retained when the dummy P picture is reproduced, the intermittent video information displayed as a still image can be created. Note that a B picture can also be used as a dummy picture. Like the dummy P picture, the dummy B picture is a picture including only information indicating that there is no change from the previous frame, that is, only information for holding the display of a frame representing a part of the video. Therefore, the P picture may be replaced with a dummy P picture, and the B picture may be replaced with a dummy B picture. Alternatively, other pictures than the GOP header and the first I picture of the GOP may be replaced with dummy B pictures.

[0057] 2. The data converter 100 of the present invention (FIG. 2)
Diagram 2A of A) shows a configuration of a data conversion apparatus 100 of the present invention for performing an image conversion process.

The data conversion device 100 includes a video / audio information input unit 20, a separation unit 22, a video conversion unit 24, a combining unit 26, and a video / audio information recording unit 28.

The MPEG-encoded video / audio information is first input to the video / audio information input unit 20, and is read into a storage work area (not shown) inside the data converter 100. The video / audio input unit 20 outputs the read MPEG video / audio information to the separation unit 22.

The separation unit 22 receives the M
The PEG video / audio information is received and separated into coded video information and coded non-video information. As described later, by using the data structure of the MPEG standard,
The process of separating video information and non-video information from MPEG video / audio information is performed in the form of digital data without decoding. The separated video information is supplied to the video conversion unit 24.
Then, the non-video information is output to the combining unit 26. Further, the separating unit 22 outputs to the video converting unit 24 information indicating how the video information and the non-video information form the MPEG video / audio information as combined information. As will be described later, the combined information is, for example, information that stores the information amount (data length) of each of the separated video information and non-video information.

The video conversion unit 24 receives the encoded video information from the separation unit 22, and converts the received encoded video information into different video information. For example, this conversion is conversion from video information representing a continuous video to video information representing an intermittent video obtained by thinning out the continuous video. Here, the intermittent video is a video obtained by thinning out a continuous video (that is, a video represented by separated video information), and thus the data amount of the intermittent video is smaller than the data amount of the continuous video. The video conversion unit 24
The combination information is corrected according to the decrease in the data amount. The video conversion unit 24 outputs the intermittent video information and the corrected combined information to the combining unit 26.

By using the data structure of the MPEG standard as described later, conversion from one video information to another different video information, including conversion from continuous video information to intermittent video information, can be performed without decoding. Performed in the form of digital data.

The combining unit 26 combines the intermittent video information received from the video conversion unit 24 and the non-video information received from the separation unit 22. At the time of combining, the combining unit 26 uses the combined information received from the video converting unit 24 to create intermittent video / audio information in which intermittent video and audio are synchronized. The created intermittent video / audio information is output to the video / audio recording unit 28.

The video / audio recording unit 28 writes the intermittent video / audio information received from the combining unit 26 into a memory or a portable terminal (not shown).

[0065] 3. The data converter 100 of the present invention (FIG. 2)
Image conversion-2B which A) is carried out is a flowchart illustrating a procedure of image conversion processing in which the data conversion device of the present invention is performed. Here, the video conversion process refers to a process of generating an intermittent video by thinning out a continuous video.

When data converter 100 (FIG. 2A) reads MPEG video / audio information via video / audio information input unit 20 (FIG. 2A) (step 201), separation unit 22
(FIG. 2A) separates the read MPEG video / audio information into video information and non-video information (step 202).
As described above, MPEG video / audio information is MPEG-encoded. Therefore, both the video information and the non-video information included in the MPEG video / audio information are encoded.
The step of separating the MPEG video / audio information into video information and non-video information may be performed without decoding the encoded MPEG video / audio information. Subsequently, the video converter 24
(FIG. 2A) generates video information different from the original video information by performing a conversion process on the separated video information (step 203). The combining unit 22 (FIG. 2A) combines again the generated video information different from the original video information and the separated non-video information (Step 204). The above-described video information conversion process includes, for example, a process of converting video information representing a continuous video into video information representing an intermittent video obtained by thinning out the continuous video. The data conversion device according to the present embodiment can perform the conversion processing and the combination processing without decoding the video information and the non-video information. The video / audio information recording unit 28 (FIG. 2A) writes the video / audio information obtained by such separation processing, conversion processing, and combination processing into a storage medium such as a memory, and a series of video conversion processing ends.

Since the data converter of the present invention does not need to decode video information and non-video information, it is possible to obtain video information without image quality deterioration. Furthermore, the load required for the calculation required for decoding can be reduced.

[0068] 4. The data converter 100 of the present invention (FIG. 2)
Description of each part of A) In the following, among the constituent elements included in the data conversion device 100 (FIG. 2A) of the present invention, particularly the main separation unit 22 (FIG.
A) The operations of the video conversion unit 24 (FIG. 2A) and the combining unit 26 (FIG. 2A) will be described in further detail.

4.1 Operation of Separation Unit 22 (FIG. 2A) FIG. 3A is a flowchart showing a procedure of a process in which the separation unit 22 (FIG. 2A) separates MPEG video / audio information into video information and non-video information.

The separation section 22 (FIG. 2A) converts the pack header 12 (FIG. 1) and the system header 13 (FIG. 1) from the MPEG video / audio information input via the video / audio information input section 20 (FIG. 2A). Read (step 301). The pack header 12 (FIG. 1) and the system header 13 (FIG. 1) do not include video information. Therefore, the separation unit 22 (FIG. 2A) writes the pack header 12 (FIG. 1) and the system header 13 (FIG. 1) as non-video information to a buffer or memory and accumulates them (Step 302). Next, the separation unit 22 (FIG. 2A) reads the packet header, that is, the packet start code 15 (FIG. 1), the packet length 16 (FIG. 1), and the synchronization information 17 (FIG. 1) from the video / audio information (step 303). The packet header also contains no video information. Therefore, the separation unit 22 (FIG. 2A) writes and stores the non-video information in the buffer (step 3).
04).

The demultiplexer 22 (FIG. 2A) reads the packet data 18 (FIG. 1) following the above-described packet header (step 305). In the packet header, the packet start code 15 (FIG. 1) includes information indicating whether the subsequent packet data is video information or audio information. Therefore, by reading the information, the separation unit 22 (FIG. 2A) can determine whether or not subsequent processing should target video information (step 306). The information included in the packet header is packet data 1
8 (FIG. 1) indicates that video information is included, the separation unit 22 (FIG. 2A) stores the packet data as video information in the buffer (step 3).
07). After that, the separation unit 22 (FIG. 2A) creates the combination information and writes it to the buffer (Step 308).

When the writing to the buffer is completed, the separation unit 22 (FIG. 2A) determines whether the processing has been performed up to the end of the video / audio information of one pack 11 (FIG. 1) (step 310). It is determined whether (FIG. 1) is the end (step 311). If video / audio information to be processed remains, the separation unit 22 (FIG. 2A) repeats the above-described processing from step 303 and step 301, respectively. If there is no video / audio information to be processed, the process ends.

The procedure of the processing performed by the separation unit 22 (FIG. 2A) has been described above.

Referring to FIG. 3B, separation section 22 (FIG. 2A)
Will be specifically described how MPEG video / audio information, that is, video information and non-video information included in the MPEG video / audio information are separated.

FIG. 3B shows a process in which MPEG video / audio information, that is, MPEG video / audio information is separated into video information and non-video information. FIG. 3B shows the input MPEG video / audio information 32 and the separated video information 30 and non-video information 34.

The MPEG video / audio information 32 includes a video data block V representing video information, a packet data (mainly audio information) block N other than video information, and a data block P containing synchronization information and attribute information. Data block P
Is information indicating a pack header, a system header, a packet start code, a packet length, and synchronization information. The number given to each data block P is a serial number for each data type. As shown, M
In the PEG video / audio information 32, video data blocks V and data blocks N and P other than the video data blocks appear alternately.

In accordance with the operation (FIG. 3A) of the separating section 22 (FIG. 2A), the MPEG video / audio information 32 includes the video data block V and the data block of the non-video information, that is, the packet data block N other than the video information and the synchronization information. And a data block P including attribute information. FIG 3B, a start position V _P of the video data blocks V of within the image information 30, (size between the starting position V _P of the video data blocks V) size between video data blocks V and V _S,
A start position N _P of a data block in the non-video information 34;
And N _S (size between the start position N _P of the data block) sizes between data blocks is shown.

As shown, the MPEG video / audio information 32 is separated into video information 30 and non-video information 34.

Next, the separation unit 22 (FIG. 2A)
08 (FIG. 3A) will be described.

FIG. 4 shows an example of the combination information. In the example shown in FIG. 4, a plurality of pieces of connection information are recorded in a table format. Binding information, video data blocks V (Figure 3B) and non-video information data blocks each start position V _P,
N _P and its sizes V _S and N _S, and the size PVS of the synchronization information of the video packet when the video is cut out.

The combination information is stored in the video data block V (FIG. 3).
B) is the information amount (data size) of the video data block V and the data block of non-video information that have been output up to that recorded when outputting B). By using the connection information, the separated video data block of the video information and the data block of the non-video information which are further thinned out can be reconnected to the video / audio information (the reconnection processing is performed by the connection unit 26 ( It will be described later in connection with the operation of FIG. 2A).

The connection information includes the data size of the synchronization information 17. The reason will be briefly described. The separation unit 22 (FIG. 2)
After being separated by A), the video converter 24 (FIG. 2A)
When the video information is thinned out, the information amount (data size) of the video information changes. Information volume (data size)
In order to re-combine the video information and the non-video information for which the video information has changed, it is necessary to correct the original packet length of the video information. This is because the amount of decrease in the information amount (data size) of the video information reduced by the thinning is subtracted from the packet length 16 to conform to the MPEG standard. When correcting the value of the packet length 16 of the video information, the position where the already received packet length 16 (FIG. 1) is stored is calculated by back calculation. In packet 14 (FIG. 1)
Data is arranged in the order of packet length 16, synchronization information 17, and packet data 18. Packet length 16 (Fig. 1)
The back calculation of the position where is stored is realized by subtracting the previous position by the data size corresponding to the information amount (data size) of the synchronization information 17 (FIG. 1) from the head position of the packet data. Here, the synchronization information 17 (FIG. 1)
Is not constant, so that scanning from the rear increases the processing load and complicates the processing. Therefore, if the data size of the synchronization information 17 is recorded in order to facilitate the back calculation of the position of the packet length 16 (FIG. 1), the load of the subsequent processing can be reduced.

By recording not only the information amount of the video information output up to that time but also the size of the data to be output in the combination information, the load of subsequent processing can be reduced.

The first line of the combination information shown in FIG. 4 is obtained when the first video data block V (FIG. 3B) is recorded.
The non-video information 34 (FIG. 3B) and the video information 30 (FIG. 3)
B), the positions N _P and V _P are both “0000”, the size N _S of the data block of the recorded non-video information 34 (FIG. 3B) is 30 bytes, and the video information 30 (FIG. 3).
Size V _S is 20 bytes of data block B), the size PVS synchronization information of the packet indicates that it is a 12 bytes.

The second line of the combination information shown in FIG.
Start position N of image information 34 (FIG. 3B) _PIs "0030",
Video data block V (1) of video information 30 (FIG. 3B)
(FIG. 3B) start position V_PIs “0020” and the record
Of the data block of the obtained non-video information 34 (FIG. 3B).
Is N_SIs 30 bytes and the video data (Fig. 3B)
Lock size V_SIs 20 bytes, packet synchronization information
Indicates that the size PVS is 3 bytes.

The third line of the combination information shown in FIG. 4 shows the same contents as described above. Here, the size N _S data blocks recorded non-video information 34 is in the size of 100 bytes, the information contained in the packet data other than the video information information (mainly voice information) This is because the size of the synchronization information of the next packet is also included in the size of the data block.

Hereinafter, every time a video packet is recorded,
Similarly, additional combination information is additionally recorded to create combination information. In this manner, the operation of repeatedly and alternately cutting out video information and non-video information according to the packet structure is M
The processing is performed until the end of the PEG video information 32 (FIG. 3B) is reached, and the combined information is created.
0 (FIG. 3B) and non-video information 34 (FIG. 3B) can be determined based on what information is being separated.

4.2 Operation of Video Converter 24 (FIG. 2A) The video converter 24 (FIG. 2A) analyzes and edits video information indicating a continuous video, and then has the same number of pictures as the input video. The intermittent video information is generated by outputting the video. That is, this operation is performed by the video converter 24 (FIG. 2).
A) is an operation of leaving only a picture having important information and replacing other frames with a dummy P picture not including information. Therefore, the output video is an intermittent video. Further, the video conversion unit 24 (FIG. 2A) corrects the combination information (FIG. 4) according to a change in the size of the video due to editing.

FIG. 5A is a flowchart showing a procedure of a process in which the video converter 24 (FIG. 2A) converts continuous video information into intermittent video information.

First, the video converter 24 (FIG. 2A) reads the sequence header from the video information (step 50).
1), the sequence header is directly written as intermittent video information (step 502). If the sequence of the video information is not at the end, the process proceeds. Next, the video conversion unit 24 (FIG. 2A) reads the GOP header from the video information (Step 504), and writes it into the GOP header of the intermittent video information (Step 505). This is the video information GO
This means that the P header is used as the GOP header of the intermittent video information. Then, the video converter 24 (FIG. 2A)
Reads an I picture from video information (step 50).
6) The I picture is written as the intermittent video information as the intermittent video information as it is (step 50).
7). If the GOP of the video information is not at the end, the process proceeds as it is (step 508). Subsequently, the video conversion unit 24 (FIG. 2A) reads the remaining pictures included in the GOP from the GOP of the video information (Step 50).
9). The reading of the picture may be a picture other than the I picture (for example, a B picture or a P picture). The video converter 24 (FIG. 2A) rejects the video information included in the read picture and writes the same number of dummy pictures as the number of the pictures as the GOP of the intermittent video information (step 510). The dummy picture is a small-sized picture that includes only information that there is no change from the previous frame. Such GOP conversion is repeated until the end of the sequence. When the end of the sequence is reached,
Returning to the reading of the sequence header (step 501), the processing is repeated. When the end of the video is reached, the conversion from the video to the intermittent video ends. If the GOP of the video information is not at the end, the sequence is not at the end, and the video information is not at the end, the processing is continued as it is.

With reference to FIG. 5B, the process in which the video converter 24 (FIG. 2A) operating as described above reads the GOP of video information and outputs the GOP of intermittent video information will be described more specifically.

FIG. 5B shows the correspondence between a certain GOP 50 extracted from the input video information and the GOP 52 of the intermittent video information obtained by converting the GOP 50. GOP5
0 is an arbitrary GOP included in the video information. GOP
Marker V _P which is attached to indicate the recorded video start position in the binding information (FIG. 4).

First, the video converter 24 (FIG. 2A)
The GOP header of 50 and the first I picture are read, and are written to the GOP 52 as they are. Then G
All the pictures after the OP50 are read.
The video information contained in the loaded picture is rejected,
Instead, it is replaced with the same number of dummy P pictures as the number of read pictures and written to the GOP 52. Dummy P
The picture is a picture containing only information that there is no change from the previous frame. The size of the dummy P picture is extremely small because it does not include video information. When this picture exists, the reproduction system continuously displays the previous frame. Such a specification of the dummy picture is defined in the MPEG standard. Due to the presence of this dummy picture, G
OP52 is reproduced like a still image of the first I picture. Further, since the GOP 52 includes the same number of pictures as the GOP 50, the reproduction time is the same as that of the GOP 50.

It should be noted here that GOP50 is
When converted to P52, the size of the dummy picture
The size of the GOP 52 is small due to its small size.
Video start position V_PIs outside GOP52
Inconsistency. So eliminate this contradiction
The video start position V_PIs included in GOP52
Need to be modified as follows. Below are some of the fixes
An example is shown. The starting position of the GOP in the video stream
S, the end position is E, and the n-th video start position V
_P(N) is V_P’(N) and V_P’(N) start
If the position is S 'and the end position is E', then V_P’(N) = (V_P(N) -S) * (E'-S ') /
By performing the calculation of (E−S) + S ′, the video start position becomes GOP5
2 can be modified to be included.

Such a series of GOP conversion operations is repeated until the end of the sequence. When the end of the sequence is reached, the process returns to reading the sequence header and repeats the processing. When the end of the video is reached, the conversion from the video to the intermittent video ends.

In the GOP conversion, the number of pictures included in the GOP 50 is the same as the number of pictures included in the GOP 52. If the number of pictures included in the GOP 50 and the number of pictures included in the GOP 52 are different, it is necessary to shift the TC (time code) value in the GOP header with respect to all subsequent GOPs, thereby increasing the processing amount.
Therefore, in the present embodiment, the GOP50 and the GOP52 have the same
There is an advantage that processing can be performed without changing the OP header.

Note that an intra picture in the middle of a GOP can be used as an intermittent video without rejection. For example, in the case where one intra picture is used for intermittent video without rejection, after replacing some pictures with a dummy P picture after the first I picture, a second intra picture is extracted and inserted, Further, the subsequent pictures are replaced with dummy P pictures. Generally GO
P is often encoded to make up a picture of about half a second. Therefore, if about one intra picture is used for one GOP, a moderate intermittent video that advances about two frames per second is obtained.

However, when it is desired to make an intermittent video to be more finely frame-advanced or when the video information itself is coded so that the number of pictures constituting the GOP is increased, not only the I picture at the beginning of the GOP but also By using an arbitrary number of intra pictures at an arbitrary position, it is possible to arbitrarily adjust the interval between frames of the intermittent video. With such flexibility, it is possible to generate an intermittent video that is displayed intermittently at a speed corresponding to the processing speed of the mobile terminal used or at a speed selected by the user. The speed according to the processing speed of the mobile terminal or the speed selected by the user is
The data is input to the data conversion device as a parameter, and the intra picture is used as each frame of the intermittent video according to the value of the parameter.

Conversely, a plurality of GOPs can be integrated into one GOP, and the interval between frames of intermittent video can be made longer than the GOP playback period.

FIG. 6 shows a plurality of GOPs included in video information.
The correspondence between a header and a picture when a single GOP 62 of intermittent video information is generated from the column 60 is shown.

The GOP column 60 is a set of GOPs obtained by extracting an arbitrary number of continuous GOPs from the original video information. For simplicity, only the first GOP and the last GOP that have been cut out are shown, and GOPs in the middle are omitted. The GOP 62 is obtained by converting the GOP sequence 60 and integrating it as a single GOP.

First, GO of the first GOP in GOP column 60
The P header information and the first I picture of the first GOP are used as the GOP of the intermittent video information. Subsequent GOP
The picture in column 60 is rejected, and a dummy P picture having no information amount is used as a picture of GOP 62 instead. At this time, the TR (temporal reference value) of each picture is corrected so as to be a serial number in the GOP 62. G after the second GOP column
The GOP header of the OP is discarded. Although the GOP header contains synchronization information such as time code, it is not always necessary, and there is no problem if it is discarded. This is, when finally reconstructed into MPEG video / audio information, MPEG
This is because the synchronization information included in the video / audio information compensates for synchronization stability.

In the created GOP 62, only the first I picture of the first GOP has video information, and the subsequent picture is a dummy P picture, which has more information than telling the reproducing system that the previous picture is continuously displayed. Absent. Therefore, when GOP62 is reproduced, the first I picture is G
It is displayed continuously during the playback period of OP62. The GOP 62 includes the same number of pictures as the GOP sequence 60. Therefore, the GOP 62 is a still image having the same length as the reproduction period of the GOP sequence 60. In this way, by integrating a plurality of GOPs into one GOP, a still frame video longer than the GOP playback period can be created.

For all GOPs included in the video,
By performing the above operation, an intermittent video can be created.

Note that GOP50 (FIG. 5B) to GOP52
(FIG. 5B), as described above, GOP62
Since the size is smaller than that of the GOP sequence 60, inconsistency occurs in the video start position included in the combination information. Therefore, the video start position held by the combination information also needs to be corrected according to the reduction in the GOP size.
An example for performing the correction will be described below. The start position of the GOP sequence 60 in the video stream is S, the end position is E, the n-th video start position N _P (n) is corrected to N _P ′ (n), and the start of the GOP 62 in the intermittent video The position is S ',
'When, N _P' end position _{E (n) = (N P} (n) -S) * (E'-S ') /
It may be calculated as (E−S) + S ′.

4.3 Operation of Combining Unit 26 (FIG. 2A) The combining unit 26 (FIG. 2A) includes the intermittent video information created by the video converting unit 24 (FIG. 2A) and the non-video separated by the separating unit 22. The information is recombined with the video and audio information based on the combination information output from the video conversion unit 53. When the intermittent video information and the non-video information are combined, it becomes intermittent video audio information. The recombination is performed with both the MPEG video / audio information and the non-video information encoded, so that the amount of calculation can be reduced.

FIG. 7A is a flowchart showing a procedure of a process in which the combining unit 26 (FIG. 2A) combines intermittent video information and non-video information.

In the combining process, the combining unit 26 (FIG. 2A) first reads the combined information from the video converter 24 (FIG. 2A) (step 701), and then reads a non-video data block from the non-video information (step 701). 702). Coupling part 2
6 (FIG. 2A) writes non-video information as video / audio information using the read combination information (step 703).
Thereafter, the combining unit 26 (FIG. 2A) determines whether the read combined information indicates the final data block of non-video information (step 704). If the combined information indicates the final data block of the non-video information, the process ends. If the combined information does not indicate the final non-video information data block, the combining unit 26 (FIG. 2A)
Reads a video data block from the video information (step 705). Thereafter, the connecting portion 26 (FIG. 2A)
The video data block of the video information is written in the video / audio information (step 706), and the process returns to the process of reading the combined information again (step 701).

Referring to FIG. 7B, a process in which intermittent video information and non-video information are combined by combining unit 26 (FIG. 2A) operating as described above will be described. FIG. 7B shows the intermittent video information 70.
And non-video information 74 and generated MPEG video / audio information 7
2 is shown. The MPEG video / audio information 72 is intermittent video information.

It is known in advance that the intermittent video / audio information 72 starts with a data block P of non-video information, and thereafter, video information, non-video information and audio information appear repeatedly and alternately, and finally ends with non-video information. .

The combining section 26 (FIG. 2A) reads the combined information, and starts the cut-out start position N _P and cut-out size N _{S of} the non-video information data block, and the cut-out start position V _P and cut-out size V _{S of the} video data block V. To get. However, at the end of the binding information, only clipping start position N _P and clipping size N _S non-video data blocks, the extraction start position V _P and clipping size V _S of the video data block information is not written Or pre-defined end code. Next, joint 2
6 (FIG. 2A), the non-video information 74, from the cutout start position N _P written to the data binding information clipping size N _S content only, cut out data blocks. For example,
The value of N _P is 0, if the value of N _S 30, the non-image information 74
The first 30 bytes are cut out from the beginning. Coupling part 26
(FIG. 2A) records the cut-out data block in the intermittent video / audio information 72. If the read combination information indicates the final non-video data block, the combination processing ends here. If the read combination information does not represent the final non-video data block, the combination unit 2
6 (FIG. 2A) is a cutout start position V from the intermittent video 70.
_A data block is cut out from _P by a cutout size V _S. For example, if the value of the value of V _P is 0, V _S is 20,
A 20-byte data block is cut out from the beginning of the video information 70 representing the intermittent video. The combining unit 26 (FIG. 2A) combines the data blocks cut out at the end of the intermittent video / audio information 72. The combining unit 26 (FIG. 2A) returns to the process of reading the combined information again, and reads the start position and the data block size of the data block to be cut out from the next non-video information 74 and the video information 70 from the combined information. Then, the combined information is read, a data block is cut out from the non-video information 74, and the data block is added to the intermittent video / audio 72. If not the last combined information, a data block is cut out from the video information 70 representing the intermittent video. The process of adding the data block to the intermittent video / audio information 72 is repeated. When the repetition operation reaches the last combined information, combined intermittent video / audio information 72 is generated.

Here, the information other than the packet length included in the packet, that is, the synchronization information among the packet start code and the synchronization information is included in the non-video information 74 when the size of the video data block does not increase. Information can be reused as it is. In general, when a video is converted to an intermittent video, the size of a data block is significantly reduced, so that it is not necessary to change the synchronization information. The packet start code is, for example, a fixed value defined in the MPEG standard. The pack header includes the synchronization information and the attribute information of the packet, but the synchronization information does not need to be changed for the same reason as the packet. The attribute information of the packet is not changed by making the intermittent video. Therefore, the header information that needs to be modified in the present invention is only the packet length 16 (FIG. 1).

The description of the configuration and operation of the data conversion device according to the first embodiment of the present invention has been completed.

(Embodiment 2) Next, a data conversion device according to a second embodiment of the present invention will be described with reference to FIG.

[0115] 1. Data conversion device 800 of the present invention (FIG.
8) Outline of operation Data conversion device 800 according to Embodiment 2 of the present invention
(FIG. 8) shows that, when a cut in which distribution of luminance, saturation, hue, or the like, that is, a change in characteristics of a video is equal to or more than a predetermined threshold in a video is included in the video, the cut is used as a still image. Generate intermittent video / audio information. The use of a cut as a still image is based on the empirical fact that the vicinity of a cut where a video changes greatly over time often contains important information on the video. By using the data conversion device according to the present embodiment, for example, in a news program, immediately after a scene is switched, a video on which a telop indicating a summary of news is displayed, or immediately after a subtitle is switched in a movie or the like, By using a video as a representative image, an outline of a scene can be grasped.

[0116] 2. Data conversion device 800 of the present invention (FIG.
8) Configuration FIG. 8 shows a data conversion device 800 including a configuration for detecting cuts contained in video.

The data conversion device 800 includes a video / audio information input unit 80, a separation unit 81, a partial decoding unit 82, a cut detection unit 83, a shot video creation and combination information correction unit 84, a combination unit 85, And a video / audio information recording unit 86.

The functions and operations of the video / audio information input unit 80, the separation unit 81, the combining unit 85, and the video / audio information recording unit 86 among the components constituting the data conversion device 800 are the same as those shown in FIG. The functions and operations of the video / audio information input unit 20, the separation unit 22, the combining unit 26, and the video / audio information recording unit 28 of the device 100 are the same as those of the apparatus 100, and thus the description thereof is omitted.

[0119] The partial encoding section 82 receives the encoded video information separated by the separating section 81, and decodes only the DC component that is a part of the video information. In MPEG, a DC (low frequency) component does not require IDCT (Inverse Discrete Cosine Transform) and can be obtained by a simple calculation. The obtained DC component of the video information is output to the cut detection unit 83 as low-resolution video information.

The cut detection unit 83 receives the low-resolution video information output from the partial decoding unit 82, and uses the received low-resolution video information (DC component of the video information) to greatly change the video per unit time. Find the part. A frame having a large change in video is called a cut frame. A cut frame can be recognized as a frame in which, for example, the distribution of luminance, saturation, hue, and the like becomes temporally equal to or greater than a certain threshold. As described above, in general, there are many frames that are imagewise important near the cut frame. The cut detection unit 83 scans the video and searches for a GOP that appears immediately after the cut. Then, the first I picture of the GOP is selected as a representative frame. The GOP including the selected representative frame and the frame number of the representative frame are output to the shot video creation and combination information correction unit 84 as cut information.

Here, the reason why the cut frame itself is not used as the representative frame is that, in the case of MPEG, the processing is easier if the video is modified in GOP units. M
When using a PEG data sequence, in order to create an intermittent video with the cut frame itself as a representative frame, M
It is necessary to decode the PEG data sequence. Decoding requires some computation. On the other hand, the GOP immediately after the cut frame
There is a trade-off in that if the first frame is a representative frame, there is a problem that an error occurs in which a less important frame is selected as the representative frame due to delay. In this embodiment, the GOP immediately after the cut frame
Will be described assuming that the first frame is a representative frame. Note that even if the amount of calculation is slightly increased, the MPEG frame is decoded, or a format other than MPEG is used. It can be used as a representative frame to represent a still image.

Shot image creation and combination information correction unit 8
5 generates an intermittent video from the video information and combination information separated by the separation unit 82 and the cut information detected by the cut detection unit 84. The shot video creation and combination information correction section 85 scans the video and searches for a representative frame recorded in the cut information. When a representative image frame is found, one intermittent video GOP extends from the GOP immediately after the representative frame to immediately before the GOP including the next representative frame.
Convert to In order to convert a plurality of GOPs included in a video into a single intermittent video GOP and correct the combined information, the same procedure as that described in connection with FIG. 6 in the first embodiment may be used. it can. By this conversion process, a still image video for continuously reproducing the first picture is created.
This still image image is called a shot image. By applying the above operation to each cut according to the cut information, a shot video corresponding to each cut is created, and the created shot videos are combined to form an intermittent image including a plurality of representative images. An image is created.

In the above description, the representative image of the intermittent video is set as the first I picture of the GOP immediately after the cut. However, in this method, since the position of the cut and the position of the representative image are shifted, the video is particularly time-sensitive. In the case of a drastic change, an image different from the representative image originally intended may be used as the representative image. Therefore, as another example of creating a shot video, a picture immediately after the cut may be used as a representative image. When the picture immediately after the cut is used as the representative image, the picture immediately after the cut is not limited to the I picture. If the picture immediately after the cut is an I picture, the I picture can be cut out and used as a representative image of the intermittent video. However, if the picture immediately after the cut is a B picture or a P picture, the preceding I picture
What is necessary is to search for a picture, decode the picture up to the picture immediately after the cut, and encode the picture immediately after the cut to generate an I picture. Accordingly, an intermittent video can be created using a representative image more faithful to the cut information as a shot video, so that an intermittent video can be created using a highly important image as a representative image. However, when such a decoding process and an encoding process are performed, the amount of calculation becomes larger than when only the I picture included in the original video information is replaced. However, the amount of calculation is small compared to the method of decoding the entire video, considering that the pictures to be decoded are only a part of the entire video and that the encoding is limited to the picture immediately after the cut. .

As another modified example of creating a shot video, an intermittent video may be created by adopting a picture after a certain period of time from the cut as a representative image since the video is likely to be disturbed immediately after the cut. . Also in this case, the representative picture may be coded into an I picture after decoding from a preceding I picture to a picture that is a fixed time after the cut, that is, a representative picture. In the shot video creation method first described in the present embodiment, a picture obtained after a lapse of a certain period of time from the cut is adopted because the leading I picture of the GOP immediately after the cut is used as the representative picture. Since this modified example also employs a picture a fixed time after the cut as the representative image,
The same effect as in the case of the first description can be obtained. However, in the case of the first description above, the picture is not necessarily a picture after a certain time has elapsed from the cut. Therefore, the picture immediately after the cut may be adopted as the representative image. As described above, the image is likely to be disturbed immediately after the cut, and the image quality of the representative image may be unstable. Therefore, a more stable image can be adopted as a representative image by decoding the preceding I picture to a picture a fixed time after the cut and encoding the representative picture, and thus a stable intermittent video can be adopted. Can be created.

In the present embodiment, the video at the time when the cut, that is, the distribution of luminance, saturation, hue, etc., becomes temporally equal to or more than a certain threshold value is adopted as the representative image. However, any information can be used as the representative image as long as the information is temporally separated from the video. Generally, when an event representing an image in which the characteristics of a video greatly changes is detected,
It can be adopted as a representative image. Event detection includes not only cut detection but also telop detection. The telop is detected as the video at the time when the total amount of the spatial gradient of the video changes greatly over time. In general, in video such as news, a summary of the scene is often presented in the video immediately after the appearance of the telop. Can be obtained. In the present embodiment, the detection of a cut has been described, but this description is also applied to the detection of an event. In order to detect an event, an event detection unit that performs the same operation as the cut detection unit 83 for detecting a cut may be provided.

The description of the configuration and operation of the data conversion device according to the second embodiment of the present invention has been completed.

It is desirable that the data converters of the first and second embodiments be realized as a device that handles video and audio information encoded according to the widely used MPEG standard. By handling information that is MPEG-standardized, the user can utilize the technology established to date with respect to the MPEG standard.

[0128]

According to the present invention, video information can be converted without decoding an encoded data sequence. Therefore, video information without image quality degradation can be obtained. Furthermore, the load required for the calculation required for decoding can be reduced.

According to the present invention, it is possible to create an intermittent video / audio having a more important image as a representative image by using an event detector for scene change of a video or the like. In addition, the intermittent image can be recorded in a form according to the mobile terminal used.

[Brief description of the drawings]

FIG. 1 is a diagram showing a general data structure (data format) of an MPEG data string.

FIG. 2A shows a data conversion device 100 of the present invention for performing a video conversion process in the first embodiment of the present invention.

FIG. 2B is a flowchart showing a procedure of a video conversion process performed by the data conversion device of the present invention.

FIG. 3A is a flowchart showing a procedure in which a separation unit 22 (FIG. 2A) separates MPEG video / audio information into video information and non-video information.

FIG. 3B is a diagram showing a process in which MPEG video / audio information (FIG. 1) is separated into video information and non-video information.

FIG. 4 is a diagram illustrating an example of combination information.

FIG. 5A is a flowchart showing a procedure of a process in which a video conversion unit 24 (FIG. 2A) converts continuous video information into intermittent video information.

FIG. 5B: GOP5 extracted from the input video information
0 and G of intermittent video information obtained by converting GOP50.
It is a figure showing the correspondence of OP52.

FIG. 6 is a diagram illustrating a correspondence relationship between a header and a picture when a single GOP 62 of intermittent video information is generated from a plurality of GOP columns 60 included in the video information.

FIG. 7A is a flowchart illustrating a procedure of a process in which a combining unit (FIG. 2A) combines intermittent video information and non-video information.

FIG. 7B is a diagram in which intermittent video information and non-video information are combined and M
FIG. 4 is a diagram illustrating a process for creating PEG video / audio information.

FIG. 8 is a diagram illustrating a data conversion device 800 including a configuration for performing cut detection included in video according to the second embodiment of the present invention.

FIG. 9 is a diagram showing a conventional data converter for recording and reproducing audio intermittently and audio continuously.

[Explanation of symbols]

 Reference Signs List 20 video / audio information input unit 22 separation unit 24 video conversion unit 26 coupling unit 28 video / audio information recording unit

 ────────────────────────────────────────────────── ─── Continuing on the front page (72) Inventor Hiroyuki Akagi 22-22, Nagaikecho, Abeno-ku, Osaka-shi, Osaka F-term (reference) 5C052 AA17 AB03 AB04 AC02 AC03 AC08 CC11 CC20 DD06 GA03 GA04 GA06 GA07 GB06 GB07 GB09 GC06 GD09 GE04 GE08 5C053 FA14 FA27 GB06 GB08 GB09 GB10 GB11 GB12 GB21 GB37 GB40 HA29 JA21 JA30 KA05 5C059 KK39 LB07 MA00 MA23 MC32 PP01 PP05 PP06 PP07 RB01 RC02 RC07 RC32 RC39 SS19 SS30 UA02 UA05

Claims (15)

[Claims] An input unit for inputting a first data string including coded first video information representing video and coded non-video information representing information other than video; The encoded first video information and the encoded non-video included in the first data sequence without decoding the first video information and the encoded non-video information. A separating unit that separates the coded first video information into a coded second video information; and a coded second video information. And a combining unit configured to combine the encoded non-video information with the separated non-video information to generate a second data string. 2. The data conversion device according to claim 1, wherein the encoded second video information represents an intermittent video obtained by thinning the video. 3. The separation unit outputs, to the video conversion unit, first combined information regarding a combination of the separated coded first video information and the coded non-video information, The video converter is configured to convert the encoded second video information from the first combined information in accordance with the conversion from the encoded first video information to the encoded second video information. Generating second combination information relating to a combination of information and the encoded non-video information, wherein the coupling unit generates the encoded second video information and the code based on the second combination information; The data conversion device according to claim 1, wherein the second data string is generated by combining the non-video information converted into the second data string. 4. The encoded non-video information synchronizes reproduction of the encoded first video information included in the first data string with reproduction of the encoded non-video information. The data conversion device according to claim 1, further comprising: synchronization information for causing the second unit to use the synchronization information as synchronization information of the encoded second data. 5. The method according to claim 1, wherein the first data string is a first MPEG data string encoded by MPEG compression, the second data string is a second MPEG data string encoded by MPEG, The coded first video information includes a predetermined number of first pictures, and the coded second video information includes the predetermined number of second pictures. Data conversion device as described. 6. The video converter, by converting a picture excluding at least one picture among the predetermined number of the first pictures into a dummy picture,
Generating the predetermined number of the second pictures from the at least one picture and the dummy picture, wherein the at least one picture is a picture including information representing a part of the video, and the dummy picture is The data conversion device according to claim 5, wherein the data conversion device is a picture including information for holding display of a part of a video. 7. The predetermined number of first pictures includes at least an intra picture and a non-intra picture, wherein the intra picture and the non-intra picture are arranged in a predetermined order, and the at least one picture The data conversion apparatus according to claim 6, wherein the data conversion apparatus includes an initially arranged intra picture. 8. The first data stream is a first MPEG data stream that has been MPEG-compressed and encoded, and the second data stream is a second MPEG data stream that has been MPEG-compressed and encoded. The encoded first video information includes a plurality of GOPs each having a plurality of pictures, the video conversion unit includes a picture excluding at least one picture of a first GOP among the plurality of GOPs, First
Generating a single GOP from at least one picture of the first GOP and the dummy picture by converting the plurality of pictures included in each of the plurality of GOPs except the GOP of The at least one picture of the first GOP is a picture including information representing a part of the video, the dummy picture is a picture including information for holding display of a part of the video, The data conversion device according to claim 1, wherein the total number of the plurality of pictures included in the plurality of GOPs is the same as the total number of pictures included in the single GOP generated by the video conversion unit. 9. Each of the plurality of GOPs is arranged in a first predetermined order, wherein the plurality of pictures included in each of the plurality of GOPs include at least an intra picture and a non-intra picture; The intra picture and the non-intra picture are arranged in a second predetermined order, and the first GOP is arranged first among the plurality of GOPs arranged in the first predetermined order. At least one picture of the first GOP is an intra picture arranged first among the plurality of pictures of the first GOP arranged in the second predetermined order. The data conversion device according to claim 8. 10. The encoded first video information includes:
A plurality of GOPs arranged in a predetermined order, each including at least an intra picture; the data conversion device; a partial decoding unit configured to decode a predetermined portion of the encoded first video information; An event detection unit that detects an event representing an image in which the characteristics of the video change significantly based on the decoded predetermined portion, wherein the video conversion unit is configured to perform at least a video conversion based on a picture corresponding to the event. 7. The data conversion device according to claim 6, wherein one picture is selected and the predetermined number of the second pictures is generated. 11. The data conversion apparatus according to claim 10, wherein the at least one picture is an intra picture of a GOP arranged next to a GOP including a picture corresponding to the event. 12. The at least one picture is a picture next to a picture corresponding to the event,
The data conversion device according to claim 10. 13. The data conversion apparatus according to claim 10, wherein the at least one picture is a picture that is played after a predetermined time has elapsed since the picture corresponding to the event was played. 14. The data conversion apparatus according to claim 7, wherein the data conversion device uses an arbitrary picture among the pictures arranged in the predetermined order as the at least one picture according to a predetermined parameter. Data conversion device. 15. The coded non-video information includes first packet length data indicating an information amount of the first video information, and the combining unit reduces the information amount of the second video information. Therefore, by modifying the first packet length data,
The data conversion device according to claim 1, wherein the second packet length data is generated and included in the second data string as the non-video information.