JP7390877B2 - Video editing device, video editing method, and computer program - Google Patents

Description

The present invention relates to a video editing device, a video editing method, and a computer program.

Traditionally, there are videos of a wide range of lengths, from videos that are a few seconds long to videos that are several hours long. Videos that are a few seconds long are easy to watch even when you are traveling on public transportation or taking a break. However, if a video is several hours long, it may not be possible to watch the entire video without interruption during the gap. Regarding sports videos, some users may only need to be able to view a digest without having to view the entire video. Therefore, in the past, many techniques for editing moving images have been proposed (for example, see Patent Documents 1 and 2).

Japanese Patent Application Publication No. 2009-182762 Japanese Patent Application Publication No. 2010-011409

However, in the conventional technology, video editing is performed by focusing on a characteristic scene (for example, a scene with a score in the case of sports). Therefore, the content may not match the user's tastes, such as not including scenes that the user really wants to see. Furthermore, although it is possible to edit a video to be shorter than the original video, there is little freedom in determining the length of the edited video. Therefore, there is room for improvement in video editing.

In view of the above circumstances, an object of the present invention is to provide a technology that can generate edited videos that better meet the needs of users.

One aspect of the present invention is a video editing unit that includes an edited video generation unit that generates edited video data having a length desired by the user and including content of a portion the user wants to view, in response to a user's request. It is a device.

One aspect of the present invention is the video editing device described above, in which the edited video generation unit receives a user's request including the length of the edited video desired by the user and the content the user wants to see, and the editing target. A model that has been trained to output video data after editing by inputting video data and index information for identifying each section obtained by dividing the video data into predetermined lengths; Then, the edited video data is generated using the video data and index information corresponding to the video data.

One aspect of the present invention is the video editing device described above, further comprising a learning section that learns the model, and the learning section learns the model based on content fed back from the user.

One aspect of the present invention is a video editing step that includes, in response to a user's request, generating edited video data that has a length desired by the user and includes content of a portion that the user wants to view. It's a method.

One aspect of the present invention is to cause a computer to execute an edited video generation step of generating edited video data having a length desired by the user and including content of a portion the user wants to view, in response to a user's request. It is a computer program for

According to the present invention, it is possible to generate an edited video that is more tailored to the needs of the user.

FIG. 1 is a diagram showing a system configuration of a video editing system in an embodiment. FIG. 1 is a schematic block diagram showing the functional configuration of a video editing device in an embodiment. It is a sequence diagram showing the flow of edited video generation processing of the video editing system in the embodiment. FIG. 7 is a diagram illustrating an example of selection items when the video that the user desires to view is baseball. FIG. 7 is a diagram illustrating an example of selection items when the video that the user desires to view is soccer. FIG. 7 is a diagram illustrating an example of selection items when the video that the user desires to view is rugby. FIG. 6 is a diagram illustrating an example of selection items when the moving image that the user desires to view is a movie. It is a sequence diagram showing the flow of update processing of the video editing system in the embodiment. FIG. 2 is a diagram for explaining a business model using a video editing system in an embodiment.

Hereinafter, one embodiment of the present invention will be described with reference to the drawings.
(overview)
An embodiment of the present invention is to generate an edited video that is more tailored to the needs of the user. In an embodiment of the present invention, the video can be edited to the length that the user wants to watch during free time such as commuting time or vacation time, and to include the part that the user wants to watch, thereby making it more tailored to the user's needs. Generate an edited video.

FIG. 1 is a diagram showing the system configuration of a video editing system 100 in an embodiment. The video editing system 100 includes a user terminal 10, a distribution server 20, a content server 30, and a video editing device 40. The user terminal 10 and the distribution server 20 are communicably connected via a network 50. Network 50 may be configured in any manner. For example, network 50 may consist of the Internet. The content server 30 and the video editing device 40 are communicably connected via a network 60. Network 60 may be configured in any manner. For example, the network 60 may be configured with a leased line. The distribution server 20 and the content server 30 are servers held by a video distribution company. A video distribution business is a business that distributes video data to users. In this embodiment, the video data may be sports video data, movie data, or video data of any object as long as it is video data.

The user terminal 10 is a communication device owned by a user who views a video. The user terminal 10 accepts requests from users. Requests from users include, for example, information that allows the user to specify the length of a video that the user wishes to watch during free time such as while traveling on public transportation or during a break, or the content of the portion of the video that the user wishes to view. Contains possible information. The information that can specify the length of the video may be, for example, a numerical value that specifies the length of time (for example, 10 for 10 minutes, etc.). The information that can specify the content of the portion of the video that the user wants to view may be, for example, information on an item directly selected by the user, or may be text information. The user terminal 10 can play the edited video data distributed from the distribution server 20. Here, the edited video data is video data generated by editing by the video editing device 40, and is a video that includes the length of the user's preference and the content of the portion that the user wants to view. The user terminal 10 is configured using an information processing device such as a tablet terminal, a smartphone, or a notebook computer.

The distribution server 20 transmits video data or edited video data to the user terminal 10 via the network 50 in response to a request from the user terminal 10 . The distribution server 20 is configured using, for example, an information processing device such as a personal computer.
The content server 30 is a server that stores video data transmitted by the distribution server 20. For example, the content server 30 stores unedited video data and edited video data. Content server 30 transmits video data to be edited to video editing device 40 via network 60. The content server 30 receives the edited video data from the video editing device 40 via the network 60.

The video editing device 40 is a device that edits video data to be edited so that the video meets the user's request. The video editing device 40 generates edited video data that reflects the preferences of each user, for example, by inputting video data to be edited and a user's request. The video editing device 40 is a device held by a different video distribution provider.
Video data may be transmitted and received between the content server 30 and the video editing device 40, and address information indicating a storage location of the video data may be transmitted and received. When transmitting and receiving address information, the content server 30 and the video editing device 40 access the destination indicated by the address information and acquire desired video data.

FIG. 2 is a schematic block diagram showing the functional configuration of the video editing device 40 in the embodiment. The video editing device 40 includes a communication section 41, a storage section 42, and a control section 43.
The communication unit 41 communicates with external devices. The communication unit 41 communicates with the distribution server 20 or the content server 30 via the network 60, for example. The communication unit 41 receives, for example, video data transmitted from the content server 30 and a user's request. The communication unit 41 transmits the edited video data to the content server 30, for example.

The storage unit 42 stores model data 421 and index files 422. By inputting the user's request, the video data to be edited, and the index file 422, the model data 421 can be created with the length desired by the user and the edited version that includes the content of the part the user wants to view. This is a model trained to output video data. The model data 421 may be different model data for each user, or may be common model data for all users.

The index file 422 is a file in which index information for identifying the content of each section is associated with each section obtained by dividing the video data into a predetermined length. If the video data is sports video data, the index information includes, for example, before the start of the game, during the game, which team is attacking or defending, which players are shown, what the play is, etc. This is information for identifying the content, such as what kind of scene it is. If the video data is of a movie, the index information is information for identifying the content, such as whether the performer is an actress or an actor, what scene or background it is, what kind of scene it is, etc. . What kind of index information is assigned to which section is set in advance by the video distribution company or the administrator of the video editing system 100. A plurality of index information may be assigned to each section. For example, a plurality of index information such as which players are shown in Team A's attack may be added to one section. Each section may be of different length. In the index file 422, information on an allowable length during editing may be set for each section. The storage unit 42 is configured using a storage device such as a magnetic hard disk device or a semiconductor storage device.

The control unit 43 is configured using a processor such as a GPU (Graphics Processing Unit) and a memory. The control unit 43 functions as an acquisition unit 431, a learning unit 432, and an edited video data generation unit 433 by executing programs.
The acquisition unit 431 acquires various information. The acquisition unit 431 acquires, for example, video data. The acquisition unit 431 acquires a user's request. The acquisition unit 431 acquires feedback information transmitted from the user terminal 10. The feedback information is information such as corrections to the edited video input by the user of the user terminal 10 after viewing the edited video.

The learning unit 432 learns model data. The learning unit 432 generates the model data 421 by learning using, for example, the index file 422, video data, video editing time, and correct answer data. The learning unit 432 updates the model data 421 by learning using the feedback information.

The edited video data generation unit 433 generates edited video data using the model data 421, the index file 422, and information requested by the user. In this way, the edited video data generation unit 433 generates edited video data that has the length desired by the user and includes the content of the portion the user wants to view, in response to the user's request.

FIG. 3 is a sequence diagram showing the flow of edited video generation processing by the video editing system 100 in the embodiment. The edited video generation process in FIG. 3 is executed when the user of the user terminal 10 requests distribution of edited video data.
The user of the user terminal 10 operates the user terminal 10 to access a video viewing site of a video distribution provider (step S101). The distribution server 20 transmits information about the video viewing site to the user terminal 10 in response to a request from the user terminal 10 (step S102). As a result, the video viewing site is displayed on the screen of the user terminal 10. Here, the user selects the type of video he or she wishes to view. On the screen of the user terminal 10, for example, a playback button for the selected video and a digest button for editing the selected video into a video of the length desired by the user are displayed. Assume that the user presses the digest button (step S103). The user terminal 10 transmits information on the selected button to the distribution server 20 (step S104).

The distribution server 20 receives button information transmitted from the user terminal 10. Since the button information includes information indicating that the digest button has been pressed by the user, the distribution server 20 transmits selection item information according to the type of video selected by the user to the user terminal 10 (step S105).
The user terminal 10 receives the selection item information transmitted from the distribution server 20. The user terminal 10 displays the received selection item information on the screen and receives selection item input from the user (step S106).

4 to 7 are diagrams showing examples of selection items displayed on the user terminal 10. FIG. 4 is a diagram illustrating an example of selection items when the video the user desires to view is baseball, and FIG. 5 is a diagram illustrating an example of selection items when the video the user desires to view is soccer. 6 is a diagram showing an example of selection items when the video the user wants to view is rugby, and FIG. 7 is a diagram showing an example of selection items when the video the user wants to view is a movie. It is a diagram. Each pattern will be explained in detail below.

First, explanation will be given using FIG. 4. As shown in FIG. 4, when the video the user wishes to view is baseball, there are items on the screen of the user terminal 10 for inputting information regarding the editing range, players, plays, and scenes. Is displayed.
The editing range item is an item for inputting the length and editing range of the edited video that the user wants to view. In FIG. 4, a viewing time item is shown as an item for inputting the length of the edited video, and a range item is shown as an item for inputting an editing range. Here, "first half", "second half", and "all games" are shown in the range item. The user inputs the desired length of viewing time (for example, any length between 1 minute and 30 minutes) in the viewing time field. The user selects one of "first half,""secondhalf," and "all games" in the range item as the range that he/she wants to watch during the free time.

The player item is an item for inputting the team name and player name that the user wants to watch. In FIG. 4, a team item is shown as an item for inputting the name of the team the user wants to watch, and a player selection item is shown as an item for inputting player names. Here, "Team A", "Team B", "...", "Both teams" are shown in the team item, and "Player A", "Player B", "..." are shown in the team item.・”, “All players” are shown. The user selects one of "Team A", "Team B", "...", and "Both Teams" in the team item. The user selects one or more of "Player A", "Player B", "...", and "All players" in the player item.

The play item is an item for inputting the attack or defense and play that the user wants to watch. In FIG. 4, an offense and defense item is shown as an item for inputting an attack or defense that the user would like to view, and a play selection item is shown as an item for inputting a play. Here, "attack" and "defense" are shown in the offensive and defensive items, and "offensive play" and "defensive play" are shown in the play selection item. The user selects one or more of attack and defense in the offense and defense item. The user selects one or more plays from among offensive plays and defensive plays in the play selection item. Note that the user terminal 10 may be configured to prevent the user from selecting the defensive play item in the play selection item when the user selects "attack" in the offensive and defensive item.

In the example shown in FIG. 4, home run, hit, strikeout, sending bunt S, sending bunt F, and no selection are shown as examples of offensive plays. Here, the suffix S represents success, and the suffix F represents failure. Note that the offensive play is not limited to this. For example, the offensive play may include any offensive play in baseball, such as squeeze S and squeeze F. In the example shown in FIG. 4, a strikeout, double play S, double play F, four-ball, and no selection are shown as examples of defensive plays. Note that the defensive play is not limited to this. For example, the defensive play may include any defensive play in baseball, such as triple play S, triple play F, and dropped balls.

The scene item is an item for inputting a scene that the user wants to view. In FIG. 4, a scene selection item is shown as an item for inputting a scene that the user wants to view. The user selects one or more scenes that the user desires to view in the scene selection item. In the example shown in FIG. 4, pinch, reversal, and no selection are shown as examples of scenes. Note that the scene is not limited to this. For example, the scene may include a walk-off, a fine play, a request, a sending off, and the like.

For example, if a user wants to watch the climax of a match focusing on batting in 10 minutes, the user can select the editing range of viewing time "10 minutes", the range "second half", the player Select teams "both teams", player selection "all players", offensive and defensive play "attack", play selection "no selection", scene selection "pinch + reversal + walk-off".
For example, if the user wants to mainly watch player A's pitching scene during the 20 minutes of the train journey, the user can select the editing range of viewing time "20 minutes" and the editing range of "all games" from the selection items shown in Figure 4. , player team "Team A", player selection "Player A", play selection "Defense", play selection "Strikeout + Double Play S", and scene selection "Pinch + Fine Play".

Next, it will be explained using FIG. As shown in FIG. 5, when the video the user wishes to view is soccer, the user terminal 10 screen displays items for inputting information regarding the editing range, players, plays, and scenes. be done. Here, since the information regarding the editing range, players, plays, and scenes is substantially the same as the information shown in FIG. 4, detailed explanation will be omitted. Only the different points will be explained below.

In the example shown in FIG. 5, shot S, shot F, through pass S, through pass F, and no selection are shown as examples of offensive plays. Note that the offensive play is not limited to this. For example, attacking plays include any offensive play in soccer such as centering S, centering F, throw-in S, throw-in F, free kick S, free kick F, corner kick S, corner kick F, etc. may be included. In the example shown in FIG. 5, Goal Save S, Goal Save F, Clear S, Clear F, and No Selection are shown as examples of defensive plays. Note that the defensive play is not limited to this. For example, the defensive play may include any defensive play in soccer, such as offside trap S and offside trap F.
In the example shown in FIG. 5, pinch, PK, and no selection are shown as examples of scenes. Note that the scene is not limited to this. For example, the scene may include a warning or the like.

For example, if a user wants to watch a score scene of a match centered around Player A and Player B of team A in 5 minutes, the user can select the viewing time of 5 minutes from the selection items shown in FIG. ”, Range “All games”, Player team “Team A”, Player selection “Player A + Player B”, Play selection “Attack”, Play selection “Shoot S + Free kick S”, Scene selection “PK”. .
For example, if the user wants to watch the highlights of a game focusing on attacks during the 30 minutes of the lunch break, the user can select the editing range of viewing time "30 minutes" and the range of "all games" from the selection items shown in Figure 5. , player teams "both teams", player selection "all players", offensive and defensive play "attack", play selection "Shoot S + Shoot F + Centering S + Centering F + Throw-in S + Throw-in F + Free kick S + Free kick F + Through pass S + Through pass F ”, select scene selection “PK”.

Next, it will be explained using FIG. As shown in FIG. 6, when the video the user wishes to view is rugby, the user terminal 10 displays items for inputting information regarding the editing range, players, plays, and scenes as selection items on the screen. be done. Here, since the information regarding the editing range, players, plays, and scenes is substantially the same as the information shown in FIG. 4, detailed explanation will be omitted. Only the different points will be explained below.

In the example shown in FIG. 6, try S, try F, scrum S, scrum F, and no selection are shown as examples of attacking plays. Note that the offensive play is not limited to this. For example, the offensive play may include any offensive play in rugby, such as penalty kick S, penalty kick F, drop kick S, drop kick F, and lineout. In the example shown in FIG. 6, intercept S, intercept F, tackle S, tackle F, and no selection are shown as examples of defensive plays. Note that the defensive play is not limited to this.
In the example shown in FIG. 6, pinch, reversal, and no selection are shown as examples of scenes. Note that the scene is not limited to this. For example, a scene may include a video determination, etc.

For example, if a user wants to view a digest of a game that focuses on tackles in 5 minutes, the user can select from the selection items shown in FIG. Select the player team "both teams", player selection "all players", play selection "defense", play selection "tackle S + tackle F", scene selection "pinch + video judgment".
For example, if a user wants to view a 10-minute digest of a set piece for research purposes, the user can select from the selection items shown in FIG. Select teams "both teams", player selection "all players", offensive and defensive play "attack", play selection "scrum S + scrum F + lineout S + lineout F", and scene selection "no selection".

Next, it will be explained using FIG. 7. As shown in FIG. 7, when the video that the user wishes to view is a movie, there are items on the screen of the user terminal 10 for inputting information regarding the editing range, performers, scenes, and scenes. Is displayed.
The editing range item is an item for inputting the length and editing range of the edited video that the user wants to view. In FIG. 7, a viewing time item is shown as an item for inputting the length of the edited video, and a range item is shown as an item for inputting an editing range. Here, "first half", "second half", and "whole" are shown in the range item. The user inputs the desired length of viewing time (for example, a desired length between 1 minute and 30 minutes) in the viewing time field corresponding to the length of the edited video. The user selects one of the first half, the second half, and the whole of the items in the range corresponding to the editing range.

The performer item is an item for inputting the gender and person of the performer that the user wishes to view. In FIG. 7, a gender item is shown as an item for inputting the gender of the performer the user wants to watch, and a person item is shown as an item for inputting a person. "A", "...", and "all performers" are shown. Here, "female", "male", and "both" are shown in the gender item. The user selects one of "female," "male," and "both" in the gender field. The user selects one or more of "actor A", "actress A", "...", and "all performers" in the person field.

The scene item is an item for inputting the background and scene that the user wants to view. In FIG. 7, a background item is shown as an item for inputting a background that the user wants to view, and a scene selection item is shown as an item for inputting a scene. The user selects one or more desired backgrounds in the background item. In the example shown in FIG. 7, "bright", "dark", "city", "sea", "mountain", "indoor", "outdoor", and "no selection" are shown as examples of the background. Note that the background is not limited to this. For example, the background may include any background that is assumed to be the background of a movie. The user selects one or more desired scenes in the scene selection item. In the example shown in FIG. 7, "action", "violence", "love", "animal", "child", and "no selection" are shown as examples of scenes. Note that the scene is not limited to this. For example, the scene may include any scene that is assumed to be a scene in a movie, such as a conversation or cheers.

The scene item is an item for inputting a scene that the user wants to view. In FIG. 7, a scene selection item is shown as an item for inputting a scene that the user wants to view. The user selects one or more scenes that the user desires to view in the scene selection item. In the example shown in FIG. 7, "synopsis", "encounter", "farewell", "pinch", "serious", and "no selection" are shown as examples of scenes. Note that the scene is not limited to this. For example, scenes may include comedies, horror, musicals, reversals, climaxes, and the like.

For example, if the user wants to view the entire synopsis in 30 minutes, the user can select from the selection items shown in FIG. Select both, character "All performers", scene background "No selection", scene selection "No selection", and scene selection "Synopsis".
For example, if the user wants to see only the meeting and farewell scene between actor A and actress A, the user can select the viewing time of "10 minutes" for the editing range and the "entire" range from the selection items shown in FIG. , performer's gender "both", person "actor A + actress A", scene background "no selection", scene selection "no selection", scene selection "encounter + farewell".

The user terminal 10 transmits information on the selection items input by the user to the distribution server 20 as input information (step S107). This allows the video to be edited. While the video editing device 40 is editing the video, that is, until the user terminal 10 receives the edited video data, the user terminal 10 displays information indicating that editing is in progress (for example, "Editing in progress"). ” is displayed on the screen. The distribution server 20 receives input information transmitted from the user terminal 10. The distribution server 20 transmits the received input information and information on the video selected by the user to the content server 30 (step S108). The content server 30 receives input information transmitted from the distribution server 20 and information on the video selected by the user. The content server 30 refers to stored content data and obtains video data corresponding to the received video information. Content server 30 transmits the acquired video data and input information to video editing device 40 (step S109).

The communication unit 41 of the video editing device 40 receives video data and input information transmitted from the content server 30. The acquisition unit 431 acquires the video data and input information received by the communication unit 41 (step S110). The acquisition unit 431 outputs the acquired video data and input information to the edited video data generation unit 433. The edited video data generation unit 433 generates edited video data using the output video data and input information, the model data 421, and the index file 422 (step S111). Specifically, the edited video data generation unit 433 first obtains index information corresponding to video data from the index file 422. Next, the edited video data generation unit 433 generates edited video data by inputting the acquired index information, input information, and video data into the model data 421. Here, if the model data 421 is different for each user, the edited video data generation unit 433 generates the target model data 421 by comparing the user ID associated with the model data 421 and the user ID of the request source. get.

The edited video data generation unit 433 transmits the generated edited video data to the content server 30 via the communication unit 41 (step S112). The content server 30 receives edited video data transmitted from the video editing device 40. The content server 30 stores the received edited video data in association with the video data before editing (step S113). The content server 30 transmits the edited video data to the distribution server 20 (step S114). The distribution server 20 receives edited video data transmitted from the content server 30. The distribution server 20 transmits the received edited video data to the user terminal 10 (step S115).

The user terminal 10 receives edited video data transmitted from the distribution server 20. The user terminal 10 stores the received edited video data. After that, the user terminal 10 plays back the saved edited video data in response to the user's operation (step S116).

FIG. 8 is a sequence diagram showing the flow of update processing of the video editing system 100 in the embodiment. The update process in FIG. 8 is executed when the user of the user terminal 10 inputs feedback information.
After viewing the edited video data, the user operates the user terminal 10 to input feedback regarding the viewing results (step S201). The user terminal 10 transmits the input feedback content to the distribution server 20 (step S202). The distribution server 20 receives the feedback content sent from the user terminal 10. The distribution server 20 transmits the received feedback content to the video editing device 40 (step S203).
The communication unit 41 of the video editing device 40 receives the feedback content transmitted from the distribution server 20. The acquisition unit 431 acquires the feedback content received by the communication unit 41 (step S204). The acquisition unit 431 outputs the acquired feedback content to the learning unit 432. The learning unit 432 learns the model using the feedback content (step S205). The learning unit 432 stores the model after learning as model data 421 (step S206).

FIG. 9 is a diagram for explaining a business model using the video editing system 100 in the embodiment.
As shown in FIG. 9, as a business model using the video editing system 100, a fixed-rate viewing contract is made between the user and the video distribution business, and the business that holds the video editing device 40 and the video distribution business In the meantime, a BtoBtoC model may be considered in which pay-as-you-go is charged according to the usage of the video editing device 40. This allows the user to view edited video data for a fixed fee.

According to the video editing system 100 configured as described above, it is possible to generate an edited video that better meets the needs of the user. Specifically, in response to a user's request, the video editing device 40 generates edited video data that has a length desired by the user and includes the content of the portion the user desires to view. Thereby, it is possible to generate video data that preferentially includes scenes that the user wants to view. Furthermore, since the edited video data is created in a length of time that the user prefers, there is a high degree of freedom. Therefore, it becomes possible to generate an edited video that is more tailored to the user's needs.

Conventionally, the following problems have arisen when editing videos.
・Video sources were limited, such as having to prepare and submit specific video files.
・There is a need to digest videos provided by various media sources and video distribution companies, but it is necessary to prepare the video source yourself.
- It is necessary for the user to process the video source intentionally to digest it (put it in a folder and start processing), which makes the operation complicated.
On the other hand, in the video editing system 100, when selecting a video that the user wants to edit, the user only has to select it from among the videos provided by the video distribution provider. Then, the user only needs to input editing details for the selected video. In this way, the user does not have to prepare the video source and the operations do not become complicated. Therefore, convenience can be improved.

Previously, in order to view the edited video, it took time and effort to move the video file. On the other hand, in the video editing system 100, edited video data can be viewed by accessing a site provided by a video distribution company. Therefore, it becomes possible to view the desired edited video data using a simple method.

The number of user terminals 10 may be plural.
Some of the functions included in the video editing device 40 may be included in another housing. For example, the learning section 432 included in the video editing device 40 may be included in another device. When configured in this way, the video editing device 40 acquires model data that has been trained by another device from the other device. The video editing device 40 stores the acquired model data in the storage unit 42. Then, the edited video data generation unit 433 generates edited video data using the model data 421 stored in the storage unit 42.

The video editing device 40 may be configured to re-edit the remaining portion in response to a user's request when the user interrupts viewing the edited video data. For example, when a user interrupts viewing 30 minutes of edited video data for 15 minutes, the video editing device 40 re-edits the remaining 15 minutes of video data in response to the user's request. When configured in this way, the video editing device 40 acquires edited video data to be re-edited and a user request from the content server 30. The user request sent here includes information that allows you to specify how many minutes of edited video data you have viewed, and information that allows you to specify how long you want to edit the remaining video data. The number of minutes the edited video data has been viewed may be determined based on the time when the user interrupted viewing. The edited video data generation unit 433 generates video data by deleting the portion viewed by the user from the acquired edited video data, and inputs the generated video data, index information, and user request into the model data 421. By doing so, edited video data after re-editing is generated.
With the above configuration, video data that the user has partially viewed can be edited into video data of a shorter duration. Therefore, it becomes possible to generate an edited video that is more tailored to the user's needs.

The video editing device 40 in the embodiment described above may be realized by a computer. In that case, a program for realizing this function may be recorded on a computer-readable recording medium, and the program recorded on the recording medium may be read into a computer system and executed. Note that the "computer system" herein includes hardware such as an OS and peripheral devices. Furthermore, the term "computer-readable recording medium" refers to portable media such as flexible disks, magneto-optical disks, ROMs, and CD-ROMs, and storage devices such as hard disks built into computer systems.

Furthermore, a "computer-readable recording medium" refers to a storage medium that dynamically stores a program for a short period of time, such as a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line. It may also include a device that retains a program for a certain period of time, such as a volatile memory inside a computer system that is a server or client in that case. Further, the above-mentioned program may be one for realizing a part of the above-mentioned functions, or may be one that can realize the above-mentioned functions in combination with a program already recorded in the computer system. It may be realized using a programmable logic device such as an FPGA (Field Programmable Gate Array).

Although the embodiments of the present invention have been described above in detail with reference to the drawings, the specific configuration is not limited to these embodiments, and includes designs within the scope of the gist of the present invention.

DESCRIPTION OF SYMBOLS 10... User terminal, 20... Distribution server, 30... Content server, 40... Video editing device, 41... Communication unit, 42... Storage unit, 421... Model data, 422... Index file, 43... Control unit, 431... Acquisition unit , 432...Learning unit, 433...Edited video data generation unit

Claims (4)

an edited video generation unit that generates edited video data having a length desired by the user and including content of a portion the user wants to view, in response to a user's request;
Equipped with
The edited video generation unit generates a user's request including the length of the edited video desired by the user and the content the user wants to see, video data to be edited, and each video data obtained by dividing the video data into predetermined lengths. A model trained to output edited video data by inputting index information for identifying sections, the user's request, the video data, and index information corresponding to the video data. A video editing device that generates the edited video data using . further comprising a learning section that learns the model,
The video editing device according to claim 1 , wherein the learning unit learns the model based on content fed back from the user. an edited video generation step of generating edited video data having a length desired by the user and including content of a portion the user wants to view, in response to a user's request;
has
In the edited video generation step, the user's request including the length of the edited video desired by the user and the content the user wants to see, the video data to be edited, and each segment of the video data divided into predetermined lengths. A model trained to output edited video data by inputting index information for identifying sections, the user's request, the video data, and index information corresponding to the video data. A video editing method that generates the edited video data using . an edited video generation step of generating edited video data having a length desired by the user and including content of a portion the user wants to view, in response to a user's request;
make the computer run
In the edited video generation step, the user's request including the length of the edited video desired by the user and the content the user wants to see, the video data to be edited, and each segment of the video data divided into predetermined lengths. A model trained to output edited video data by inputting index information for identifying sections, the user's request, the video data, and index information corresponding to the video data. A computer program for generating the edited video data using .

Images