KR20220160316A - Method for generating customized video based on objects and service server using the same - Google Patents

Abstract

The present application relates to a method for generating a customized video for each object and a service server using the same. A method for generating a customized video for each object according to an embodiment of the present invention may include the steps of: analyzing an original video and identifying at least one object appearing in each frame image included in the original video; cropping the object from the frame image and generating a cropped image corresponding to each of the objects; and collecting the cropped images for each object and generating a customized video for each object.

Description

Method for generating customized video based on objects and service server using the same}

The present application relates to a method for generating a custom video for each object capable of automatically generating a custom video for each object appearing in a video, and a service server using the same.

Recently, with the spread of networks such as the Internet and the high functionality of various devices (personal computers, smart phones, cameras, etc.) used by general users, video services that provide various services desired by users are widely used.

In addition, with the popularity of K-POP and the growth of idol fandom culture, the number of playbacks of idols' stage videos among video services is increasing, and recently there has been a demand to intensively watch performances of specific members, such as member-specific direct cams. It is increasing. In response, major broadcasting stations are also providing video services, such as separately producing and providing direct cams created by separately filming each member.

However, in the case of direct cams for each member, there is a limitation in that a person must separately shoot each member's direct cam, or separately edit and upload each member's video from the original video already filmed.

An object of the present application is to provide a method for generating a custom video for each object capable of automatically generating a custom video for each object appearing in a video, and a service server using the same.

An object of the present application is to provide a method for creating a customized video for each object and a service server using the same, which can create an optimally customized video by automatically switching to the screen in which the corresponding object is best when photographed with a plurality of cameras.

An object of the present application is to provide a method for generating a video customized for each object capable of displaying a video customized for each object in various user experiences (UX) and a service server using the same.

A method for generating a custom video for each object according to an embodiment of the present invention relates to a method for generating a custom video for each object of a service server, by analyzing an original video, and generating at least one frame image included in the original video. identifying the object; cropping the object from the frame image to generate a cropped image corresponding to each of the objects; and collecting the cropped images for each object and generating a customized video for each object.

A service server according to an embodiment of the present invention includes an object identification unit for analyzing an original video and identifying at least one object appearing in each frame image included in the original video; an image cropping unit that crops the objects from the frame image and generates cropped images corresponding to the respective objects; and an editing unit that collects the cropped images for each object and creates a customized video for each object.

In addition, the solution to the above problem does not enumerate all the features of the present invention. Various features of the present invention and the advantages and effects thereof will be understood in more detail with reference to specific embodiments below.

According to the method for generating a custom video for each object according to an embodiment of the present invention and the service server using the same, it is possible to automatically generate a custom video for each object appearing in the video. Therefore, real-time processing of original images or processing of a large amount of original images is possible.

According to the method of generating a video customized for each object according to an embodiment of the present invention and the service server using the same, it is possible to automatically switch to the screen on which the corresponding object is best displayed among a plurality of target videos, thereby providing optimal fit for the corresponding object. You can create a video.

However, the effects that can be achieved by the method for generating customized videos for each object and the service server using the same according to embodiments of the present invention are not limited to those mentioned above, and other effects not mentioned above can be seen from the description below. It will be clearly understood by those skilled in the art to which the invention belongs.

1 is a schematic diagram showing a system for generating a customized video for each object according to an embodiment of the present invention.
2 is a block diagram showing a service server according to an embodiment of the present invention.
3 is an exemplary view showing an original video according to an embodiment of the present invention.
4 is an exemplary diagram illustrating a video customized for each object according to an embodiment of the present invention.
5 is a schematic diagram illustrating object recognition in a frame image according to an embodiment of the present invention.
6 is a schematic diagram showing a cropped image according to an embodiment of the present invention.
7 and 8 are schematic diagrams illustrating display of an original video and a customized video in a user terminal according to an embodiment of the present invention.
Fig. 9 is a schematic diagram showing shooting of an original video according to an embodiment of the present invention.
10 is a block diagram showing a service server according to another embodiment of the present invention.
11 is a flowchart illustrating a method for generating a custom video for each object according to an embodiment of the present invention.

Hereinafter, the embodiments disclosed in this specification will be described in detail with reference to the accompanying drawings, but the same or similar components are given the same reference numerals regardless of reference numerals, and redundant description thereof will be omitted. The suffixes "module" and "unit" for components used in the following description are given or used together in consideration of ease of writing the specification, and do not have meanings or roles that are distinct from each other by themselves. That is, the term 'unit' used in the present invention means a hardware component such as software, FPGA or ASIC, and 'unit' performs certain roles. However, 'part' is not limited to software or hardware. A 'unit' may be configured to reside in an addressable storage medium and may be configured to reproduce one or more processors. Thus, as an example, 'unit' refers to components such as software components, object-oriented software components, class components and task components, processes, functions, properties, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays and variables. The functionality provided within the components and 'parts' may be combined into a smaller number of elements and 'parts' or further separated into additional elements and 'parts'.

In addition, in describing the embodiments disclosed in this specification, if it is determined that detailed descriptions of related known technologies may obscure the gist of the embodiments disclosed in this specification, the detailed descriptions thereof will be omitted. In addition, the accompanying drawings are only for easy understanding of the embodiments disclosed in this specification, the technical idea disclosed in this specification is not limited by the accompanying drawings, and all changes included in the spirit and technical scope of the present invention , it should be understood to include equivalents or substitutes.

1 is a schematic diagram showing a system for generating a customized video for each object according to an embodiment of the present invention.

Referring to FIG. 1 , a system for generating customized videos for each object according to an embodiment of the present invention may include a user terminal 1 and a service server 100 .

The user terminal 1 may execute various types of applications, and may provide a user with a visual or auditory display of the running applications. The user terminal 1 may include a display unit for visually displaying an application, and may include an input unit receiving a user's input, a communication unit, a memory storing at least one program, and a processor.

The user terminal 1 may be a mobile terminal such as a smart phone or a tablet PC, and may also include a fixed device such as a desktop according to embodiments. Specifically, the user terminal 1 includes a mobile phone, a smart phone, a laptop computer, a digital broadcasting terminal, a personal digital assistant (PDA), a portable multimedia player (PMP), a slate PC, A tablet PC, an ultrabook, a wearable device (eg, a smartwatch, a smart glass, a head mounted display (HMD)), etc. may be included. .

The user terminal 1 may access an App store or a play store to download and install various applications. Depending on the embodiment, it is also possible to download through wired or wireless communication with the service server 100 or other devices (not shown).

Meanwhile, the user terminal 1 may be connected to the service server 100 through a communication network. Here, the communication network may include a wired network and a wireless network, and specifically, various local area networks (LAN: Local Area Network), metropolitan area networks (MAN: Metropolitan Area Network), wide area networks (WAN: Wide Area Network), etc. may include networks. Also, the communication network may include the well-known World Wide Web (WWW). However, the communication network according to the present invention is not limited to the networks listed above, and may include a known wireless data network, a known telephone network, a known wired or wireless television network, and the like.

Among applications executed in the user terminal 1, a video application may be included, and the user terminal 1 may play a video provided from the service server 100 using the video application. Here, the service server 100 may provide video in various ways such as real-time streaming and VOD (Video on Demand).

The service server 100 may be connected to the user terminal 1 through an application and may provide various online services in response to requests from the user terminal 1 . Here, the service server 100 may provide a video service for providing various videos to the user terminal 1. In this case, the service server 100, in response to the request of the user terminal 1, the database (not shown), etc. The stored original video OV may be provided to the user terminal 1 . In addition, depending on the embodiment, after the service server 100 identifies each object appearing in the original video OV, a customized video may be generated for each object, and according to a request of the user terminal 1 It is also possible to provide a customized video of a corresponding object to the user terminal 1 .

For example, as shown in FIG. 3, the service server 100 may collect performance videos of various musicians as an original video (OV), and a plurality of members (M1, M2, M3, M4). , M5), as shown in FIG. 4, the original video (OV) can be edited to create custom videos (PV1, PV2, PV3, PV4, PV5) for each member.

In other words, in the case of performance videos such as idols, the demand for intensive viewing of a specific member's performance is gradually increasing. We provide services such as: However, in the related art, a customized video for each member is created by separately shooting a customized video for each member directly using a human force or separately editing a video for each member from an entire original video (OV) that has already been filmed. In this case, since manpower is limited, it is difficult to extract a customized video from the original video OV in real time or to quickly process a large amount of the original video OV.

On the other hand, according to the service server 100 according to an embodiment of the present invention, since a customized video corresponding to each object can be automatically generated from the collected original video (OV), the customized video is processed in real time. I can easily carry out mass processing. However, although a musician's performance video has been described as an example, the present invention is not limited thereto, and the service server 100 according to an embodiment of the present invention may generate customized videos from original videos in various fields. For example, it is possible to create a customized video for a specific player or position from a relay video of a sports game, or to create a customized video for a specific person or vehicle from a security video captured by a surveillance camera. Hereinafter, the operation of the service server 100 according to an embodiment of the present invention will be described with reference to FIG. 2 .

Referring to FIG. 2 , the service server 100 according to an embodiment of the present invention may include an object identification unit 110, an image cropping unit 120, an editing unit 130, and a transmission unit 140. .

The object identification unit 110 may analyze the original video OV to identify objects appearing in each frame image included in the original video OV. The original video OV may include a plurality of frame images, and the original video OV may indicate movement by continuously displaying the plurality of frame images according to a set frame rate. Accordingly, the object identification unit 110 may extract frame images from the received original video OV, and may identify objects included in the frame image by performing image processing on the extracted frame images. Depending on the embodiment, it is also possible for the object identification unit 110 to capture the original video OV at an arbitrary frame rate and generate respective frame images.

Also, the object identification unit 110 may detect objects included in each frame image, and may distinguish the detected object from other objects. Specifically, the object identification unit 110 may learn features of objects to be extracted, such as shape or luminance, in advance using a machine learning algorithm, and utilize this to detect objects included in each frame image. . For example, the object identification unit 110 may repeatedly learn the shapes of various people using a machine learning algorithm, and through this, the shapes of people included in the frame image may be extracted as objects. At this time, the object identification unit 110 may utilize machine learning algorithms such as CNN, RNN, PCA (Principal Component Analysis), Logistic Regression, and Decision Tree.

Then, as shown in FIG. 5, the object identification unit 110 may specify the extracted objects as bounding boxes (b1, b2, b3) and display them in the frame image fi1, A corresponding object may be identified by comparing feature points shown in the boxes b1, b2, and b3. At this time, the object identification unit 110 may generate identification information for the identified object, and the identification information includes an identification ID (Identification) for distinguishing objects, coordinate information of a bounding box, reliability of an object identification result, etc. this may be included.

Depending on the embodiment, first, the object identification unit 110 distinguishes objects in the first frame image fi1 of the original video OV using the respective bounding boxes b1, b2, and b3, and each has a different identification ID. can be given first. Thereafter, with respect to the remaining frame images, feature points of objects in the current frame image are compared with feature points of objects in the previous frame image to confirm identity, and then objects whose identity is recognized are given the same identification ID, respectively. objects can be identified.

The image cropping unit 120 may crop objects from the frame image to generate cropped images corresponding to the respective objects. 5 and 6, the object identification unit 110 may create bounding boxes b1, b2, and b3 corresponding to each object in the frame image fi1, and the image cropping unit 120 may generate cropped images ci1, ci2, and ci3 by cropping the frame image fi1 based on the bounding boxes b1, b2, and b3.

6 shows that the image cropping unit 120 generates cropped images ci1, ci2, and ci3 by cropping the frame image fil to the same width and height as the bounding boxes b1, b2, and b3. Depending on , it is also possible to cut with a margin of a certain interval from the bounding boxes (b1, b2, b3). Furthermore, the image cropping unit 120 may generate cropped images ci1, ci2, and ci3 by cropping the frame image fi1 and enlarging or reducing the frame image fi1 according to a set standard.

The editing unit 130 may collect the cropped images for each object and create a customized video for each object. Here, since the cropped images are generated for each object, if the cropped images for each object are collected in chronological order, a customized video corresponding to the object can be created. That is, as shown in FIG. 4 , the editing unit 130 collects the cropped images of the respective members M1, M2, M3, M4, and M5, so that the customized videos PV1, PV2, PV3, and PV4 of the individual members are combined. , PV5) can be created. Here, since each of the customized videos (PV1, PV2, PV3, PV4, and PV5) includes only the dancing figure of a specific member, it is possible for the user to intensively watch only the movement of a desired member using the corresponding customized video.

Meanwhile, in the case of objects continuously appearing during the playing time of the original video OV, the number of cropped images generated by the image cropping unit 120 is equal to the number of frame images included in the original video OV. Accordingly, the editing unit 130 may collect the cropped images to create a customized video having the same length as the original video OV.

However, in some cases, some objects may disappear and appear within the original video (OV) for a while. In this case, some cropped images corresponding to the frame image are missing, resulting in out-of-synchronization between the original video (OV) and the customized video. problems may arise. In this way, when the cropped image corresponding to the frame image is missing, the editing unit 130 duplicates the previous cropped image or includes an arbitrary image including a phrase indicating that the corresponding object does not appear, The original video (OV) and custom video can be synchronized.

The transmission unit 140 may transmit the generated customized video to each user terminal 1 . The user terminal 1 may request an original video (OV) or a customized video from the service server 100, and the transmission unit 140 provides the original video (OV) or a customized video according to the request of the user terminal 1. can do.

As shown in FIG. 4, when a plurality of customized videos (PV1, PV2, PV3, PV4, and PV5) are generated, the transmission unit 140 transmits a plurality of customized videos (PV1, PV2, PV3, PV4, and PV5). After arranging, the customized videos (PV1, PV2, PV3, PV4, PV5) can be set to be simultaneously displayed on the screen of the user terminal 1 and transmitted.

In addition, as shown in FIG. 7 , the transmission unit 140 may display only the personalized video (PV) of the object corresponding to the selection input when a user's selection input for any one of a plurality of objects is applied. In this case, it is also possible to set and transmit the original video (OV) together with the customized video (PV) of the object corresponding to the selection input to be displayed at the same time.

Depending on the embodiment, when a user selects one customized video in a state where a plurality of customized videos (PV1, PV2, PV3, PV4, PV5) are displayed as shown in FIG. 4, the selected customized video (PV ) and the original video (OV) can be set to be displayed at the same time.

In addition, as shown in FIG. 8, the transmission unit 170 may transmit a personalized video (PV) to be displayed. That is, within the original video (OV), a region corresponding to the customized video (PV) and a region not corresponding to the customized video can be set to be visually distinguished and displayed. For example, a region corresponding to a personalized video (PV) may be highlighted, and other regions not corresponding to a video for each object may be displayed in a manner of shading.

Additionally, an embodiment in which the transmission unit 140 transmits the customized video PV in a different display method according to the resolution of the original video OV is also possible. For example, when the resolution of the original video (OV) is greater than or equal to a set value, the customized video (PV) can be set to be displayed in a separate area distinct from the original video (OV) as shown in FIG. 7 . On the other hand, if the resolution of the original video (OV) is less than the set value, as shown in FIG. 8, it is also possible to set the customized video (PV) to be visually distinguished and displayed within the original video (OV).

Meanwhile, depending on the embodiment, there may be cases in which the same objects are simultaneously photographed in different photographing environments using a plurality of cameras. That is, as shown in Figure 9, a plurality of cameras (C1, C2, C3, C4) can be installed in order to photograph the stage performance of idols, etc., each camera is a plurality of images taken at different illumination and angle Target videos can be created.

In this case, the service server 100 according to an embodiment of the present invention collects each of the target videos generated by the plurality of cameras C1, C2, C3, and C4, and displays the objects M1, M2, and M3. You can create a custom video for At this time, the service server 100 may automatically switch to a screen on which the corresponding object appears best among a plurality of target videos to generate an optimally customized video.

Specifically, the object identification unit 110 may be provided with an original video, and at this time, the original video may include all of a plurality of target videos captured by the respective cameras C1, C2, C3, and C4. Accordingly, the object identification unit 110 may first synchronize a plurality of target videos, and extract frame images of the same viewpoint from the synchronized target videos. At this time, the object identification unit 110 may synchronize the target videos based on sound sources included in the target videos. For example, since the shooting start points of the cameras C1, C2, C3, and C4 may be different, they may not be accurately synchronized by matching the start points of the target videos. Accordingly, when the same sound source is recorded in each target video, such as an idol's stage performance, synchronization can be performed based on the sound source. Then, the object identification unit 110 may identify each object from the extracted frame images.

The image cropping unit 120 may generate cropped images for each target video by cropping each of the objects included in the frame images of the same viewpoint.

The editing unit 130 may select a cropped image that satisfies setting conditions from among cropped images generated for each target video, and create a customized video for the object. That is, since cropped images representing the same object are generated for each target video, it is necessary to select one of the plurality of cropped images. Here, in order to select an optimal cropped image, setting conditions may be prepared in advance, and the editing unit 130 may select a cropped image according to the setting conditions.

Specifically, the setting conditions may include a size ratio occupied by the cropped image in the frame image, brightness of the cropped image, whether or not image distortion occurs in the cropped image, and the like. That is, if the cropped image is too small in the frame image, a problem such as a lower resolution may occur when the cropped image is enlarged to create a customized video. Accordingly, the editing unit 130 may select a cropped image occupying a large size ratio within the frame image.

In addition, if the cropped image is too dark, the object included in the cropped image is difficult to see, so the editor 130 may set a setting condition to select a relatively bright cropped image.

In addition, since it is difficult to recognize an object in a cropped image when it is shaken or out of focus, the editor 130 may set a setting condition to preferentially select a cropped image in which image distortion does not occur.

In this way, using the setting conditions, the editing unit 130 can select an optimal cropped image to create a customized video for the object.

Additionally, depending on the embodiment, when the editing unit 130 selects a cropped image extracted from one target video, a customized video may be set by selecting cropped images extracted from the same target video for at least a set number of frame images. have. That is, target videos selected within one custom video can be prevented from being converted too frequently.

In addition, although the case where the setting conditions are determined in advance is exemplified here, depending on the embodiment, it is also possible to set which cropped image to select using a machine learning technique such as deep learning.

Meanwhile, as shown in FIG. 10, the service server 100 according to an embodiment of the present invention may include a physical configuration such as a processor 10 and a memory 40, and the memory 40 may include one or more modules configured to be executed by the processor 10. Specifically, one or more modules may include an object identification module, an image cropping module, an editing module, a transmission module, and the like.

The processor 10 may execute various software programs and instruction sets stored in the memory 40 to perform various functions and process data. The peripheral interface unit 30 may connect input/output peripheral devices of the image masking device 100 to the processor 10 and the memory 40, and the memory controller 20 may connect the processor 10 or the service server 100. When a component accesses the memory 40, it may perform a function of controlling memory access. Depending on embodiments, the processor 10, the memory controller 20, and the peripheral interface unit 30 may be implemented on a single chip or may be implemented as separate chips.

The memory 40 may include a high-speed random access memory, one or more magnetic disk storage devices, non-volatile memory such as a flash memory device, and the like. In addition, the memory 40 may further include a storage device located away from the processor 10 or a network attached storage device accessed through a communication network such as the Internet.

Here, as shown in FIG. 10, the service server 100 according to an embodiment of the present invention includes an operating system in the memory 40, an object identification module corresponding to an application program, an image cropping module, an editing module, and A transmission module may be included. Here, each module is a set of instructions for performing the above-described functions, and may be stored in the memory 40 .

Accordingly, in the service server 100 according to an embodiment of the present invention, the processor 10 may access the memory 40 and execute instructions corresponding to each module. However, since the object identification module, image cropping module, editing module, and transmission module perform operations corresponding to the above-described object identification unit 110, image cropping unit 120, editing unit 130, and transmission 140, respectively. , A detailed description is omitted here.

11 is a flowchart illustrating a method for generating a custom video for each object according to an embodiment of the present invention. Here, each step of FIG. 11 may be performed by a service server according to an embodiment of the present invention. Hereinafter, with reference to FIG. 11, a method for generating a custom video for each object according to an embodiment of the present invention will be described.

First, the service server may analyze the original video and identify objects appearing in each frame image included in the original video (S110). The service server may pre-learn features of objects to be extracted, such as shape or luminance, using a machine learning algorithm, and may utilize this to detect objects included in each frame image. In this case, the service server may display the extracted objects in a frame image by specifying them as a bounding box, etc., and then compare feature points appearing in the bounding box to identify the corresponding object. At this time, the service server may generate identification information for the identified object. The identification information may include an identification (ID) for distinguishing objects, coordinate information of a bounding box, reliability of an object identification result, and the like. The service server may compare the previous frame image, etc., and set an identification ID for each object.

Thereafter, the service server may crop the objects from the frame image to generate cropped images corresponding to the respective objects (S120). Here, the service server may crop the frame image based on the bounding box, and may generate a cropped image by cutting a margin of a predetermined interval from the bounding box according to an embodiment. In addition, it is also possible to create a cropped image by cropping the frame image and enlarging or reducing it according to the set standard so that the object in the customized video is clearly displayed.

When the cropped image is generated, the service server may collect the cropped images for each object and create a customized video for each object (S130). Here, since the cropped images are generated for each object, if the cropped images for each object are collected in chronological order, it is possible to generate a customized video corresponding to the object. Therefore, it is possible to easily create a customized video that can focus only on the movement of a specific object, unlike an original video in which motions of all objects can be generally checked.

After generating the customized video, the service server may transmit the customized video to the user terminal in response to a request from the user terminal (S140). Depending on the embodiment, a plurality of generated customized videos may be arranged, configured to display the customized videos simultaneously on the screen of the user terminal, and then transmitted. In addition, when a user's selection input for any one of a plurality of objects is applied, it is also possible to display only a customized video of an object corresponding to the selection input. In this case, the original video may be displayed simultaneously with the customized video of the object corresponding to the selection input.

In addition, the service server may visually divide and display a region corresponding to the customized video and a region not corresponding to the customized video within the original video. For example, a region corresponding to a customized video may be highlighted, and the rest of the region not corresponding to a video for each object may be displayed in a manner of shading.

Additionally, it is possible for the service server to transmit the customized video in a different display method according to the resolution of the original video. For example, if the resolution of the original video is greater than the set value, the custom video is displayed in a separate area distinct from the original video, and if the resolution of the original video is less than the set value, the custom video is visually distinguished within the original video. It is also possible to set it to display.

Meanwhile, according to an embodiment of the present invention, there may be cases in which a plurality of target videos in which the same objects are simultaneously photographed in different shooting environments are included in an original video. In this case, the service server may automatically switch to a screen on which the corresponding object is best displayed among a plurality of target videos to generate an optimally customized video.

Specifically, the service server may first synchronize a plurality of target videos, and extract frame images of the same viewpoint from the synchronized target videos. Depending on the embodiment, the service server may synchronize target videos based on sound sources included in the target videos. Thereafter, the service server may identify each object from the frame images of the same viewpoint.

The service server may crop each of the objects included in frame images of the same view, and may generate cropped images for each target video.

In addition, the service server may generate a customized video for the object by selecting a cropped image that satisfies setting conditions from among cropped images generated for each target video. That is, since cropped images representing the same object are generated for each target video, it is necessary to select one of the plurality of cropped images. Here, in order to select an optimal cropped image, setting conditions may be prepared in advance, and the service server may select a cropped image according to the setting conditions.

Specifically, the setting conditions may include a size ratio occupied by the cropped image in the frame image, brightness of the cropped image, whether or not image distortion occurs in the cropped image, and the like. That is, it is possible to select an optimal cropped image for each object using setting conditions, and through this, a customized video for the object can be created.

The above-described present invention can be implemented as computer readable code on a medium on which a program is recorded. The computer-readable medium may continuously store programs executable by the computer or temporarily store them for execution or download. In addition, the medium may be various recording means or storage means in the form of a single or combined hardware, but is not limited to a medium directly connected to a certain computer system, and may be distributed on a network. Examples of the medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical recording media such as CD-ROM and DVD, magneto-optical media such as floptical disks, and ROM, RAM, flash memory, etc. configured to store program instructions. In addition, examples of other media include recording media or storage media managed by an app store that distributes applications, a site that supplies or distributes various other software, and a server. Accordingly, the above detailed description should not be construed as limiting in all respects and should be considered illustrative. The scope of the present invention should be determined by reasonable interpretation of the appended claims, and all changes within the equivalent scope of the present invention are included in the scope of the present invention.

The present invention is not limited by the foregoing embodiments and accompanying drawings. It will be clear to those skilled in the art that the components according to the present invention can be substituted, modified, and changed without departing from the technical spirit of the present invention.

1: user terminal 100: service server
110: object identification unit 120: image cropping unit
130: editing unit 140: transmission unit

Claims (16)

In the method of generating a custom video for each object of a service server,
analyzing the original video and identifying at least one object appearing in each frame image included in the original video;
cropping the object from the frame image to generate a cropped image corresponding to each of the objects; and
Collecting the cropped images for each object and generating a custom video for each object.
The method of claim 1 , wherein identifying the object comprises:
Detecting objects included in the frame image to create a bounding box, comparing feature points in the bounding box to identify each object, and generating identification information for the object. How to create a custom video.
The method of claim 1, wherein the identification information
and at least one of identification (ID) of the objects, coordinate information of the bounding box, and reliability of an identification result of the object.
According to claim 1,
The custom video generation method for each object, characterized in that it further comprises the step of transmitting the custom video to a user terminal.
The method of claim 4, wherein the transmitting step
When a plurality of custom videos are created, the custom videos are arranged, and the custom videos are set to be simultaneously displayed on one screen and transmitted.
The method of claim 4, wherein the transmitting step
When a selection input for any one of a plurality of objects is applied, the original video and a custom video of an object corresponding to the selection input are set to be simultaneously displayed and transmitted.
According to claim 1,
The original video includes a plurality of target videos in which the same objects are simultaneously photographed in different shooting environments,
Identifying the object
The method of generating a customized video for each object, characterized in that by synchronizing the target videos, extracting frame images of the same viewpoint, and identifying each object from the frame images.
8. The method of claim 7, wherein generating the cropped image
and generating cropped images corresponding to objects included in the frame images for each target video.
8. The method of claim 7, wherein identifying the object comprises:
A method for creating custom videos for each object, characterized in that the target videos are synchronized based on the sound sources included in the target videos.
The method of claim 8, wherein generating the customized video
The method of generating a customized video for each object, characterized in that by selecting a cropped image that satisfies a set condition among the cropped images generated for each target video, and generating a customized video for the object.
11. The method of claim 10, wherein the setting condition is
The method of generating a custom video for each object, comprising at least one of a size ratio occupied by the cropped image in the frame image, brightness of the cropped image, and whether or not image distortion occurs in the cropped image.
The method of claim 4, wherein the transmitting step
The method of generating a customized video for each object, characterized in that in the original video, a region corresponding to the customized video and a region not corresponding to the customized video are visually separated and displayed, and then transmitted.
13. The method of claim 12, wherein the transmitting step
In the original video, a region corresponding to the customized video is highlighted and an area not corresponding to the customized video is sound-translated.
The method of claim 4, wherein the transmitting step
If the resolution of the original video is equal to or greater than the set value, set the custom video to be displayed in a separate area distinct from the original video;
If the resolution of the original video is less than a set value, the customized video creation method for each object, characterized in that setting to visually distinguish and display the customized video within the original video.
A computer program stored in a medium in combination with hardware to execute the method of generating a custom video for each object of any one of claims 1 to 14.
an object identification unit that analyzes the original video and identifies at least one object appearing in each frame image included in the original video;
an image cropping unit that crops the objects from the frame image and generates cropped images corresponding to the respective objects; and
A service server comprising an editing unit that collects the cropped images for each object and creates a customized video for each object.

Images