KR100269809B1 - Apparatus for generating object of a video stream in the digital video editing system - Google Patents

Abstract

PURPOSE: A device for generating a video stream object in a digital video edition system is provided to generate a video stream object systematically and generally configured with various media, based on a main stream moving picture. And the device is enabled to be appropriate for a digital video edition. CONSTITUTION: A media object generator(201) reads media configuring a video stream from a database, to extract temporal and spatial characteristics. A synchronizing object generator(202) designates reproducing time and reproducing length of generated each media object, using the temporal characteristics of the generated media. A spatial configuration object generator(203) designates spatial relations of each media object and generates a spatial scale object, using the spatial characteristics of the generated media. And a video stream object generator(204) generates a static picture, a moving picture, and a video stream object for text and sound, using the generated temporal scale object and the spatial scale object.

Description

Video Stream Object Generator in Digital Video Editing System

The present invention relates to a digital video editing system based on multimedia object modeling, and more particularly, to an apparatus for generating a video stream object to systematically and comprehensively edit various media based on moving images.

The representative video editing method is video algebraic editing. The video algebra system proposed by Weiss et al. Provides four functions: creation, compositing, output and description. The creation function creates an input video for editing, and the output function outputs multiple video streams simultaneously on the screen using multiple windows. Annotation is added, and compositing is the most important function in video algebra systems, and it changes the temporal characteristics by operating on one or more video data. Operators for composition include continuous, sum, intersection, difference, iteration, expansion, and restriction.

As described above, the existing video editing method can compose one video by changing several videos or change the contents of the video. However, only the video-oriented editing method is proposed. Therefore, it was not possible to edit a video stream including various media such as voice, text, graphic video and animation.

In order to improve this problem, a multimedia data editing method has been proposed. The representative editing methods are Hamakawa and Rekimoto's method and a hierarchical multimedia data model. Hamakawa and Rekimoto's method proposed a multimedia data model for organizing and playing objects. And for the construction of multimedia data objects, we defined objects called temporal glue and box. Temporal glue is an object that controls the length of an object and has three properties: normal length, extended length, and reduced length. We also use boxes to represent the spatial arrangement of multimedia objects. For playing multimedia data, context class, media class, and viewer class are defined.

The hierarchical multimedia data model is composed of four hierarchical structures including the definition layer, the processing layer, the presentation layer, and the control layer of the multimedia data. Processing and editing can be performed. In the definition layer of data, classes for representing objects of multimedia data are defined. The processing layer represents the temporal characteristics of objects. Objects with temporal characteristics are called events, and for this purpose, operators such as Concatenate, Overlay, Extent, and Insert are defined. The presentation layer defines user interface elements such as spatial layouts, output formats, windows, and icons between objects and events, and the control layer defines interactions for users and input / output devices.

However, since these methods are for editing multimedia documents including video, they cannot be directly used for editing digital video, which is composed of various media on a video base, and does not support spatial editing such as video composition. Do not. This can be clearly seen in the comparison diagram between the multimedia document and the digital video shown in FIG.

That is, as shown in FIG. 1A, since a multimedia document is generally based on text, which is a non-continuous medium, it is less time-limited and relative location information between the media is important. Most media except texts are represented by icons, and when a user clicks on an icon, a corresponding medium is represented. On the contrary, since digital video is based on moving picture, which is a continuous medium, as shown in FIG. 1B, it is subject to a lot of time constraints, and it is important to synchronize the media. In addition, media except for sound exist in the moving picture, and its position is indicated by the coordinates in the moving picture. Due to such a difference, the existing method of editing a multimedia document cannot be used as it is for digital video editing.

An object of the present invention is to provide an apparatus for generating a video stream object suitable for digital video editing, in order to solve the above problems.

Another object of the present invention is to provide a video stream object generating apparatus for generating a video stream object systematically and comprehensively composed of various media based on a mainstream video.

According to an aspect of the present invention, there is provided a video stream object generating apparatus comprising: a media object generating unit for extracting temporal and spatial features by reading respective media constituting a video stream from a database; A synchronization object generation unit for designating a reproduction time and a reproduction length of each of the media objects generated using the temporal characteristics of the media generated from the media object generation unit, and generating a time scale object; A spatial constructing object generating unit for designating a spatial relationship of each of the generated media objects by using spatial features of the media generated from the media object generating unit, and generating a spatial scale object; And a video stream object generator for generating a video stream object for a still image, a video, a sound, and a character by using the time scale object generated from the sync object generator and the space scale object generated from the spatial component generator. It features.

1 is a comparison diagram of a multimedia document and a digital video,

2 is a block diagram of a video stream object generating apparatus according to the present invention;

3 is a hierarchical structure diagram of a media object according to the present invention;

4 is a hierarchical structure diagram of a configuration object according to the present invention;

FIG. 5 is a diagram illustrating an adjustment of a time scale object made in the synchronization object generator illustrated in FIG. 2.

<Description of the code | symbol about the principal part of drawing>

201: media object generation unit 202: synchronous object generation unit

203: Spatial composition object generator 204: Video stream object generator

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

2 is a block diagram of an apparatus for generating a video stream object according to the present invention, which includes a media object generator 201, a sync object generator 202, a spatial configuration object generator 203, and a video stream object. The generation unit 204 is configured.

The video stream object generator configured as described above is driven as follows.

First, media objects represent temporal and spatial characteristics of moving images, still images, texts, and sounds, which are media constituting the video stream, and are divided into temporal objects and spatial objects. The temporal object is an object representing the temporal characteristics of the temporal medium and has information on the reproduction length and the reproduction speed. The spatial object expresses the spatial characteristics of the medium having visual information and has information on the area where the visual medium is represented on the screen. Video object has the property of temporal object and spatial object, and still image object and text object have the property of spatial object only. Sound objects only have properties of time objects.

The media object includes a region of interest (ROI) object representing an area of interest or an important part of each medium. The region of interest of the temporal object is represented by a time interval object and the region of interest of the spatial object is represented by a rectangular object. The region of interest object is used to adjust the play length or presentation region of each medium. The region of interest of the temporal object is the part that can characterize the content of the target video. This region is designated by the automatic search by the user interface or image processing at the time of video input, and the editor builds the database with these objects. . Once the region of interest has been determined, other data is designated as the secondary region. The hierarchy of such media objects is shown in FIG. 3.

Therefore, the media object generation unit 201 reads each media constituting the video stream from a database (not shown) and extracts temporal and spatial features. The media object generator 201 is implemented as a Gen function as shown in Table 1 below.

Rectangle = Gen (Width, Height); Interval = Gen (Start, End); SpatialROI = Gen (Position, Rectangle); TemporalROI = Gen (Interval); SpaceMedia = Gen (Rectangle, SpatialROI); TimeMedia = Gen (Interval, TemporalROI ;; MovingPicture = Gen (Name, SpaceMedia, TimeMedia, MainStream); Sound = Gen (Name, TimeMedia); StillPicture = Gen (Name, SpaceMedia); Text = Gen (Content, Point, Font, Color, Rectangle);

That is, in order to create a media object, first, a rectangular object, a time interval object, a location object, and an ROI object must be created. The created time object and the region of interest object are used to create a time object and a space object, and through this, a video object, a still image object, and a sound object are created. And the text object is created by inputting the content, font, color and size to the text generator (not shown).

The synchronization object generator 202 generates an object corresponding to the synchronization object in the hierarchical structure of the structure shown in FIG. 4 as shown in Table 2, wherein the synchronization object is the playback time and playback length of each medium based on the frame of the main stream video. Indicates.

for (id = 0; id <(# ofMedia-1); id ++) {SyncObj = Synchron (Media, StartFrame, Length); ScaleObj = TScale (TimeMedia, Method); ScaledSyncObj = Compose (SyncObj, ScaleObj); }

In Table 2, the Synchron function specifies the frame of the mainstream video where a particular medium starts playing, specifies the playback length, and the Tscal function specifies how to adjust the playback length of each time medium, and synchronizes with the Compose function. An object is created, and this process is performed for all media except mainstream video. Therefore, when the media object is generated from the media object generator 201, the playback time and the playback length of the generated media objects are designated and a time scale object is generated.

This time scale object has a function of actively adjusting the temporal characteristics of each medium so as not to violate the constraints of the mainstream video. There are two ways to control the origin and region of interest. The method of reference to the origin is to act on the time medium to adjust based on the starting point of the medium, and the region of interest method is to adjust the region of interest to be expressed first.

Referring to FIG. 5, a method of adjusting a scale of a time object is illustrated. FIG. 5A illustrates a video having a playback length and a playback speed of 10 seconds and 10 frames per second (Frames / Second), and the shaded portion is a 60th frame and a 100th frame. Area of interest (between 6 and 10 seconds). When the medium having a length of 10 seconds shown in FIG. 5A is reduced to a length of 7 seconds or extended to a length of 13 seconds, FIG. 5B is a result of adjusting the reproduction length to 7 seconds or 13 seconds based on the origin, and FIG. 5C is a region of interest. This is the result of adjustment based on. When adjusting based on the origin, the region of interest is not considered at all, but the region of interest is preferentially expressed in the region of interest method. Accordingly, the sync object generator 202 generates a sync object for sound, still image, moving picture, and text as shown in the hierarchical structure shown in FIG.

The spatial configuration object generation unit 203 specifies the spatial relationship between the spatial media objects and generates the spatial scale objects through the process shown in Table 3.

for (id = 0; id <(# ofSpaceMedia-1); id ++) {MoveObj = Moving (Direction, Velocity); SpaceCompObj = SRelation (SpaceMedia, Position, Rectangle, Priority, MoveObj); ScaleObj = SScale (Method); ScaledSpaceCompObj = Compose (SpaceCompObj, ScaleObj); }

In other words, Move object is created by Moving function and SRelation function is to set position, size and priority of each spatial object. The SScale function specifies how to create a spatial scale object. The SScale function creates a spatial configuration object by using the Compose function. This process is performed on spatial objects except main stream video.

The spatial object is one of the constituent objects like the synchronization object described above. Therefore, the configuration object expresses the spatial relationship between the synchronization object and the spatial object described above. The spatial composition object specifies the position and size of the region where the spatial media is represented in the main stream video, and determines the order in which the spatial media are preferentially displayed on the screen. This priority is expressed as an integer, and the lower the number, the lower the priority. A coordinate system is required to represent a spatial component and defines a coordinate system whose origin is the upper left endpoint of the main stream video.

The spatial object includes a move object, which indicates a moving direction and a moving speed of the area in which the spatial medium is displayed, and can change the position of the spatial medium during playback time through the moving object. The moving direction has one value among the up, down, left and right, and the moving speed is expressed in the number of pixels per frame. In addition, scale objects exist in the spatial composition object, which controls the size of the region where the spatial medium is represented.

Accordingly, as can be seen in the hierarchical structure diagram of the constituent objects shown in FIG. 4, the spatial constituent object generator 203 generates spatial constituent objects for still images, moving images, and text.

The video stream object generator 204 generates a bitstream object as an object generated by the spatial component generator 203 and the sync object generator 202 as shown in Table 4.

Video Stream = Compose (ScaledSyncObj, ScaledSpaceCompObj);

As described above, the present invention can not only provide a digital video editing system with high scalability and reusability by generating a video stream object used in the digital video editing system in an encapsulated form of all interrelated information. This has the effect of shifting the editing approach from a functional decomposition approach to a data decomposition approach. In addition, each object has not only structural information such as the type and size of the medium, but also information such as the method and type of playback of the object, so that multimedia data is no longer dependent on the playback module of the system. It also has the effect of being dependent. Thus, an efficient editing and playback base can be provided.

In addition, there is an advantage in that information on changes caused by updates, modifications, and additions of media objects that exist independently can be interchanged, thereby efficiently managing related information between objects, and systematically maintaining the relationships between objects.

Claims (4)

In the digital video editing system, A media object generating unit for extracting temporal and spatial features by reading respective media constituting the video stream from a database; A synchronization object generator configured to designate a reproduction time and a reproduction length of each of the media objects generated by using the temporal characteristics of the media generated from the media object generator, and to generate a time scale object; A spatial constructing object generating unit for designating a spatial relationship of each of the generated media objects by using the spatial features of the respective media generated from the media object generating unit, and generating a spatial scale object; And And a video stream object generator configured to generate a video stream object for a still image, a video, a text, and a sound by using the temporal scale object generated from the sync object generator and the spatial scale object generated from the spatial component generator. And a video stream object generating apparatus. The method of claim 1, wherein the spatial composition object generation unit designates the position and size of the region represented in the main stream video, and specifies the spatial relationship between the media objects by determining the order in which the spatial media are preferentially expressed on the screen. And a video stream object generating apparatus. The method of claim 1, wherein the spatial object generating unit comprises a moving object for indicating the moving direction and the moving speed of the area in which the spatial medium is displayed, and changes the position of the spatial medium during the playback time through the moving object. Video stream object generator. The method of claim 1, wherein the temporal scale object and the spatial scale object actively adjust temporal and spatial characteristics of respective media using a method of origin and region of interest so as not to violate the constraints of the mainstream video. A video stream object generator, characterized by providing a function.

Images