JP2001167109A - Method for constituting group of feature descriptions of audio video information - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method for constituting a group of the feature descriptions of audio video information, which collects the feature descriptions from plural pieces of audio video information in accordance with the specific kind of feature information so as to express the descriptions as a group of feature descriptions efficiently and clearly. SOLUTION: Feature description parts 1 and 2 give the feature descriptions based on the plural kinds of feature information to individual audio video information a1 and a2. Feature description extracting parts 3 and 4 extract the feature descriptions given by the parts 1 and 2 from the information a1 and a2 based on the set specific kind of feature information (c). A feature description group constitution part 5 constitutes the group of the feature descriptions by using the extracted plural feature descriptions. A feature description group file generating part 6 describes the group of feature descriptions as the file of the feature description group. Consequently, the part 5 can collect the feature descriptions from the plural audio video information in accordance with the specific kind of feature information.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a method for constructing a feature description group of compressed or uncompressed audio / video information,
In particular, by collecting feature descriptions of a plurality of audio-video information according to a specific feature information type, and configuring a plurality of feature descriptions conforming to the feature information type as one feature description group, a plurality of audio-video program The present invention relates to a feature description group configuration method for audio / video information capable of acquiring a feature description group based on a specific feature information type.

[0002]

2. Description of the Related Art The feature description of audio / video information is currently described by MPEG (International Organization for Standardization) in MPEG.
-7 (Motion Picture Coding Experts Group phase 7
). MPEG-7 standardizes a content description and a description language for enabling efficient retrieval of compressed or uncompressed audio / video information.

[0003] In MPEG-7, feature descriptions from various viewpoints are standardized. Among the items, for example, a summary (summary) description enabling high-speed and efficient browsing of audio-video information is included. Is MPEG-7
Only deals with a single audio-video information,
The configuration and description of summary information in various summary types are possible. As the summary type, important events in the program, important audio clips, video clips, and the like are used.

For example, as shown in FIGS. 10 (a) and 10 (b), in a single audio video program 50, in other words, in closed audio video programs 50 and 51, various summary types, for example, "Home run", "Scoring scene", "Stealing scene", "Strikeout scene", ...
Can be described as a summary group.

[0005]

Among the characteristic descriptions relating to the conventional audio / video information, for example, as for the summary description, as described above, only the configuration and description of the summary information for a single audio / video program are possible. The structure and description of summary information on a plurality of audio-video programs are not specified at this time.

[0006] Further, when trying to describe a feature description group using the feature description of a summary group from a plurality of programs in the currently specified framework, for example, a plurality of programs 50, 51 as shown in FIG. When trying to describe a feature description group using the feature description of the summary group from
For example, it is expected to be as shown in FIG. That is, it is conceivable that the summary information of the summary group of each program is simply collected and described.

[0007] Therefore, the description of the conventional feature description group becomes redundant, and when searching for a desired summary from the summary group, useless processing is performed and the search time becomes longer. In addition, it is difficult to clearly describe the specification of the program to which each summary is referred. In addition, when searching for a desired summary from the summary group, it is difficult to represent a plurality of summary types in combination.

An object of the present invention is to solve the above-mentioned problem of the prior art, collect feature descriptions from a plurality of audio-video programs according to a specific feature information type, and efficiently and clearly describe descriptions as a feature description group. An object of the present invention is to provide a method of constructing a feature description group of audio / video information that can be expressed. Another object of the present invention is to provide a method for constructing a feature description group of audio / video information, which enables a desired feature description to be obtained from a feature description group by combining a plurality of feature information types.

[0009]

In order to achieve the above-mentioned object, the present invention provides a method for constructing a feature description group of compressed / uncompressed audio / video information based on a plurality of feature information types for each audio / video program. A feature description, extracting the feature description from a plurality of audio-video programs based on a specific feature information type, and forming a feature description group using the feature descriptions from the extracted plurality of audio-video programs. The first feature is that the feature description group can be described as a feature description group file.

Further, according to the present invention, as an element for describing the above-described feature description group as a feature description group file, a feature information type for forming the feature description group and its contents are described at a higher level. A second feature is that the identifier of the audio-video program to be referred to and the section designation of each feature information in the audio-video program are described at a lower level.

[0011] Further, according to the present invention, the characteristic information is a summary (outline) of the audio-video information, the characteristic information type is a summary type, and the summary description given to each audio-video program is converted to a specific summary type. A third feature is that a summary group is constructed using a plurality of summary descriptions extracted from the plurality of audio-video programs based on the extracted summary descriptions, and the summary group is described as a summary group file.

Further, according to the present invention, as an element for describing the above-mentioned summary group as a summary group file, a summary type for constituting the summary group and its contents are described at a higher level, and the audio-video referred to by each summary description is described. A fourth feature is that the program identifier and the section designation of each summary description in the audio-video program are described at a lower level.

Further, according to the present invention, the feature description types based on different feature information types are described by nesting the feature information types for constituting the feature description groups and the contents of the feature information types. The fifth feature lies in that the configuration and description of a feature description group based on the same configuration or different content of the same feature information type can be described.

Further, according to the present invention, by summarizing and describing the summary types for constituting the above-mentioned summary group and their contents and nesting them, the configuration of the summary groups based on different summary types or the same A sixth feature is that the configuration and description of the summary group based on the summary type content that differs depending on the summary type are enabled.

According to the first to sixth features of the present invention, feature descriptions belonging to each of a plurality of audio / video programs are collected according to a specific feature information type, and a feature description group is configured to efficiently and clearly Can be described. In addition, it is possible to search for desired audio / video information from a feature description group by combining a plurality of feature information types.

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail with reference to the drawings. In the present embodiment, a feature description group related to a summary for quickly and efficiently grasping the outline of the audio / video information among the feature description groups of the audio / video information will be described.

FIG. 1 shows a processing flow showing the configuration of an embodiment of the present invention. First, in the feature description units 1 and 2, feature descriptions relating to individual audio / video information a1, a2 (audio / video information 1, audio / video information 2,...) Are made based on various types of feature information.
The feature description files b1 and b2 are generated for each. At this time, each audio / video information may be compressed or uncompressed, and may be mixed in the whole.

The feature description files b1, b2 (feature description file 1, feature description file 2,...) Obtained from the plurality of pieces of audio / video information are stored in a feature description extraction unit 3,
The feature description extracting units 3 and 4 extract corresponding feature descriptions d1 and d2 from the respective feature description files b1 and b2 based on a certain feature information type c. Here, the feature information type c to be extracted may be designated by an external input from the user, or the feature description may be extracted based on all the feature information types described in each feature description file. . The feature description group forming unit 5 includes a plurality of feature descriptions d1 thus extracted.
, D2, and sends the feature description group e to the feature description group file generation unit 6. The feature description group file generator 6 forms a description as a feature description group file using the description method according to the present invention, and generates a feature description group file f.

FIG. 2 shows a feature description group e achieved by the present invention.
1 shows a specific example. In this example, the characteristic information type c corresponds to the summary type for each of the audio-video information a1 and a2, and shows an example of a summary description based on a certain summary type (main event / home run). . These audio-video program groups 21, 2
2,... (Program 1, Program 2,...), A summary description is collected based on the target summary type, and a summary group 25 is formed.

Conventionally, as shown in FIG. 10, for each audio / video program (closed audio / video program), only a summary based on various summary types (major events, major objects, etc.) can be described. However, in the present invention, according to a certain summary type,
It is possible to collect summary descriptions from a plurality of audio-video programs 21, 22,... And configure and describe a summary group 25.

FIGS. 3 and 4 show a feature description group described using a conventional feature description method and a feature description group constructed according to the present invention. As shown in FIG. 3A, in the conventional feature description group 31, the identifier 32a of the audio-video program referred to by each feature information,
32b,... Are described at the highest level, and the lower level describes the type and content of the characteristic information and the section designation of the audio / video information corresponding to the characteristic information. When browsing the feature description group 31, the feature description group file described in this manner is input to the audio-video information browsing system and interpreted. For example, the feature description of “feature information type 1” 33a, 33b,. When the user wants to browse the group, the summary based on the “feature information type 1” 33a, 33b,.
Since there is no means for determining whether or not the feature description file is described in the program represented by..., The entire feature description group file 31 must be interpreted from beginning to end. In addition, if the effective range of the reference program to which each feature information type belongs is unclear and there are many feature information types, it may be difficult to specify the feature information type.

On the other hand, in the feature description group 35 according to the present invention shown in FIG.
Are described at the highest level, and below that, the identifier 3 of the audio / video program referred to by each feature information based on the feature information type 33, 34,.
2a, 32b,... And their section designations are described.
Therefore, when it is desired to browse a feature description group based on a specific feature information type / content, for example, “feature information type 1” 33, only the highest level needs to be interpreted, and this does not conform to the desired feature information type / content. In this case, the next feature information type 34 is skipped, and if a desired feature description group is searched, the interpretation can be terminated.

Also, the reference programs 32a, 32b,... Are included for each of the characteristic information types 33, 34,. further,
In the existing feature description group 31, there are two descriptions (feature information type 1) (33a, 33b), but in the feature description group according to the present invention, there is only one description. Can be avoided, and the file size of the feature description group can be reduced. Fig. 4 (a), (b)
Shows the same contents as in FIGS. 3 (a) and 3 (b) in a table format, and a description thereof will be omitted.

FIG. 5 shows the configuration and processing flow when the feature information type c in FIG. 1 is set to “summary type”. In this specific example, a summary description is made by the summary description units 11 and 12 for each of the audio-video programs a1 'and a2'. The summary description extraction units 13 and 14 are configured to output the summary description files b1 ′, b1 ′ obtained by the summary description units 11 and 12, respectively.
b2 'is extracted from each of the summary description files b1' and b2 'according to a specific summary type (major event / home run, ...) c'. The summary composing section 15 collects the summary descriptions d1 'and d2' to compose a summary group e ', and the summary group file generating section 16
Generates a summary group file f 'using the summary group description method according to the present invention.

FIG. 6 shows, similarly to FIG. 3, a description of a feature description group using a conventional feature description method (FIG. 6 (a)).
FIG. 2B shows a feature description group constructed according to the present invention (FIG. 2B).

In the summary group file 41 according to the present invention, by setting the summary type c ′ to “summary type: main event, content: home run” in FIG. Then, by setting the summary type c ′ to “summary type: main event, content: double-bat”, the second summary group 43 is obtained from the summary group forming unit 15. Summary group file generator 16
Edits the first and second summary groups 42 and 43 into a summary group file 41 and outputs it. By the above operation, the summary group file 41 of FIG. 6B can be obtained.

FIG. 7 is an explanatory diagram of another embodiment of the present invention. FIG. 1A shows a feature description group described using a conventional feature description method, and FIG. 1B shows a feature description group described by the method of the present invention.

As shown in FIG. 2A, in the existing feature description group, the program identifier is described at the highest level, and the types and contents of the respective feature information are all described in parallel at the same level. You. Therefore, in the case of this description, it is difficult to extract a desired feature information type by combining a plurality of feature information types / contents.

On the other hand, in the feature description group of the present invention shown in FIG. 3B, the types and contents of the characteristic information are described collectively, and the different types and contents of the characteristic information are nested. It is possible to create a feature description group based on feature information types or the same feature information types and different feature information type contents.

FIG. 8 is an explanatory diagram when the “feature information type” in FIG. 7 is set to “summary type”. FIG. 1A shows a feature description group described using a conventional feature description method, and FIG. 1B shows a feature description group described by the method of the present invention.

As shown in FIG. 2A, in the existing summary group description, the program identifier is described at the highest level, and each summary type and content are described in parallel at the same level. . Therefore, in the case of this description, it is difficult to extract a desired summary description by combining a plurality of summary types / contents.

On the other hand, in the summary group description of the present invention shown in FIG. 6B, different summary types and contents are described collectively, and different summary types and contents are nested, so that different summary types are described. Alternatively, a summary description based on the different summary type contents for the same summary type is possible.
For example, in the illustrated example, the “main event” 45 and the “main objects” 46a and 46b are described as nested as summary types.

FIG. 9 is a flowchart showing an outline of the operation of FIG. 5 according to the present embodiment. In step S1,
It is determined whether or not an instruction of the nested structure configuration has been given to the summary group configuration unit 15. If the determination is negative, the operation of creating the summary group file of FIG. 3B described above is performed. On the other hand, if the judgment is affirmative, the routine proceeds to step S2, where the parent summary type c 'is set. In step S3, the summary description extraction units 13 and 14 send AV (audio video) programs 1 and 2
A summary description corresponding to the parent summary type c ′ is extracted. In step S4, the child summary type c 'is set. In step S5, the summary description extraction units 13 and 14 extract the summary description corresponding to the child summary type c' from the AV programs 1 and 2. Step S
In 6, the summary type contents are nested based on the extracted summary description. In step S7, it is determined whether or not all the summary types have been set. If the determination is negative, the process returns to step S2 and returns to step S2.
The operation of S6 is repeated. As described above, one or more nested summary groups are formed, and step S
If the determination in step 7 is affirmative, the process proceeds to step S8, where the summary group file generation unit 16 generates a summary group file as shown in FIG.

With such a nested structure, summaries based on a plurality of different types and contents of summaries can be efficiently described, and more sophisticated search and browsing of audio / video information can be performed.

[0035]

As is apparent from the above description, according to the present invention, feature descriptions from a plurality of audio-video programs are collected according to a specific feature information type. The description can be expressed efficiently and clearly. Further, it becomes possible to acquire a desired feature description from a feature description group by combining a plurality of feature information types.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a schematic configuration and a processing flow of an embodiment of the present invention.

FIG. 2 is a conceptual diagram of a summary group configured by a feature description group configuration unit in FIG. 1;

FIG. 3 is a diagram showing an example of description contents of a feature description group file obtained by the conventional method and the method of the present invention.

FIG. 4 is a diagram showing an example of description contents of a feature description group file obtained in a conventional manner and a method of the present invention in a table format.

FIG. 5 is a block diagram showing a schematic configuration and a processing flow when the feature information type is a summary type in FIG. 1;

FIG. 6 is a diagram showing an example of description contents of a summary group file obtained by the conventional method and the method of the present invention.

FIG. 7 is a diagram showing another example of description contents of a feature description group file obtained by the conventional method and the method of the present invention.

FIG. 8 is an explanatory diagram in the case where “feature information type” in FIG. 7 is set to “summary type”.

FIG. 9 is a flowchart illustrating an operation of forming a nested summary group file.

FIG. 10 is an explanatory diagram of a summary group generated by a conventional method.

[Explanation of symbols]

1, 2, ... feature description section, 3, 4, ... feature description extraction section, 5 ... feature description group configuration section, 6 ... feature description group file generation section, 1
1, 12: summary description section, 13, 14: summary description extraction section, 15: summary group configuration section, 16: summary group file generation section.

 ──────────────────────────────────────────────────続 き Continuing from the front page (72) Inventor Hiromasa Yanagihara 2-1-15 Ohara, Kamifukuoka-shi, Saitama Prefecture Inside Cadedy Research Institute, Inc. (72) Akio Yoneyama 2-1-15 Ohara, Kamifukuoka-shi, Saitama Corporation Inside the K.D. Institute (72) Inventor Haruhisa Kato 2-1-15, Ohara, Kamifukuoka-shi, Saitama F.Term in the K.D. Institute (Reference) 5B075 ND16 NK06 NR12

Claims (6)

[Claims] 1. A method for constructing a feature description group of compressed / uncompressed audio / video information, wherein each audio / video program is provided with a feature description based on a plurality of feature information types, and based on the specific feature information type. Audio-video information extracting a feature description from a plurality of audio-video programs, constructing a feature description group using the extracted plurality of feature descriptions, and describing the feature description group as a feature description group file Method of constructing a feature description group. 2. The method according to claim 1, wherein said characteristic information type is a type of summary (summary), and the summary description assigned to each audio / video program is a specific summary type. A plurality of audio-video programs based on the plurality of audio-video programs, a summary group is constructed using the plurality of extracted summary descriptions, and the summary groups are described as a summary group file. Configuration method. 3. The feature information description group configuration method for audio / video information according to claim 1, wherein the feature description group is described as a feature description group file, and a feature information type for configuring the feature description group;
The content of the feature information type is described at a higher level, and an identifier of an audio video program referred to by each feature description and a section designation of each feature information in the audio video program are described at a lower level. A method for constructing a feature description group of audio / video information. 4. The method according to claim 3, wherein when the characteristic information is a summary of the audio / video information, the summary group is described as an element for describing the summary group as a summary group file. The summary type to be composed and the contents of the summary type are described at a higher level, and at a lower level, an identifier of an audio-video program referred to by each summary description and a section designation of each summary in the audio-video program And a description method of a feature description group of audio / video information. 5. The feature description group construction method for audio / video information according to claim 1, wherein a feature information type for constituting the feature description group and the contents of the feature information type are collectively described and nested. Thus, the configuration of a feature description group based on different feature information types,
Alternatively, a feature description group configuration method for audio / video information, wherein the feature description group can be configured and described based on the content of the same feature information type and different feature information types. 6. The method according to claim 2, wherein a summary type for constituting the summary group and the contents of the summary type are collectively described and nested. A feature description group configuration method of audio / video information, wherein a configuration of a summary group based on different summary types or a configuration and description of a summary group based on different summary type contents of the same summary type is enabled.