JP2022173865A - Play analysis device, play analysis method, and, computer program - Google Patents

Abstract

To allow a play video of a sport to be viewed from any view point.SOLUTION: A play analysis device analyzing a play of a sport comprises: a processor which on the basis of play information including information on the plays of players generated on the basis of a plurality of captured images captured by at least two cameras, generates a play video being a video which reproduces the play using the three-dimensional model of the player and plays the play video from a view point instructed by a user; and an output interface which outputs the played play video to a viewing terminal.SELECTED DRAWING: Figure 11

Description

The present disclosure relates to a play analysis device, a play analysis method, and a computer program.

In recent years, by analyzing three-dimensional information of a subject from multi-viewpoint images captured by multiple cameras with different viewpoints and generating free-viewpoint video according to such three-dimensional information, free viewpoints (for example, positions and directions) can be obtained. ), a technology for generating a free-viewpoint video is known, which allows the user to observe the video of the subject. For example, three-dimensional information of the space is acquired using multiple cameras, the trajectory of the ball in a predetermined period is specified, and one or more people are within a predetermined distance from the position of the ball for a predetermined period of time or more in the predetermined period. player, set the range that includes the trajectory of the ball and the trajectory of one or more players as the range of video output, furthermore, include the set range and map it onto the spherical surface centered on the center of gravity of that range A system is known that accepts the arrangement of viewpoints and generates a video from an arbitrary viewpoint (see Patent Document 1).

JP 2018-182566 A

Conventional technology refers to tracking information corresponding to a specific scene, narrows down the players involved in the ball from the specified movement trajectory of the ball, and sets an area that includes the ball and the movement trajectory of the narrowed down players as a region of interest, Support information for viewpoint selection is provided for such an area of interest. Therefore, the conventional technology allows viewing of video from a free viewpoint in a limited local scene, but does not take into consideration the desire to view a free viewpoint video such as a player's viewpoint designated in advance or a bird's-eye view of the whole.

An object of the present disclosure is to provide a technology that allows viewing of sports play videos from any viewpoint.

A play analysis device according to an aspect of the present disclosure is a play analysis device that analyzes a sports play, and includes a plurality of players' plays generated based on a plurality of captured images captured by at least two cameras. a processor that generates a play video, which is a video that reproduces the play using the three-dimensional model of the player, based on play information including information about and an output interface for outputting the played play video to a viewing terminal.

A play analysis method according to an aspect of the present disclosure is a play analysis method for analyzing a sports play by an information processing device, and includes a plurality of images generated based on a plurality of captured images captured by at least two cameras. based on the play information including information about the player's play, using the 3D model of the player to generate a play video that is a video that reproduces the play, and reproduces the play video from the viewpoint instructed by the user. and display it on the viewing terminal.

A computer program according to an aspect of the present disclosure is a computer program for analyzing a sports play, and is a computer program for analyzing a play of a plurality of athletes generated based on a plurality of captured images captured by at least two cameras. based on the play information including information about the player, generates a play video that is a video that reproduces the play using the three-dimensional model of the player, reproduces the play video from the viewpoint instructed by the user, and reproduces the play video from the viewing terminal make the computer do what you want it to display.

In addition, these generic or specific aspects may be realized by a system, device, method, integrated circuit, computer program or recording medium, and any of the system, device, method, integrated circuit, computer program and recording medium may be implemented. may be implemented in any combination.

According to the present disclosure, a sports play video can be viewed from any viewpoint.

Schematic diagram showing a configuration example of a play analysis system according to the present embodiment Block diagram showing a configuration example of a play analysis device according to the present embodiment A diagram showing a configuration example of play information according to the present embodiment Diagram for explaining the player's posture model Flowchart showing an example of play information generation processing according to the present embodiment Flowchart showing an example of play video generation processing according to the present embodiment Schematic diagram showing a display example of play video Schematic diagram showing an operation example for a play video Schematic diagram showing a display example of the viewpoint movement button Schematic diagram showing an example of action marker filtering in the seek bar Schematic diagram showing an example of a player selection screen

Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings as appropriate. However, more detailed description than necessary may be omitted. For example, detailed descriptions of well-known matters and redundant descriptions of substantially the same configurations may be omitted. This is to avoid unnecessary verbosity in the following description and to facilitate understanding by those skilled in the art. It should be noted that the accompanying drawings and the following description are provided to allow those skilled in the art to fully understand the present disclosure, and are not intended to limit the subject matter of the claims.

(this embodiment)
<Configuration of play analysis system>
FIG. 1 is a schematic diagram showing a configuration example of a play analysis system according to this embodiment.

The play analysis system 1 is a system that analyzes an image (hereinafter referred to as a “captured image”) 121 (see FIG. 2) that captures the play of a ball game, and analyzes the movement trajectory of the moving object used in the ball game and the movement of the player. be. In this embodiment, volleyball, which is one of the ball games, will be described as an example. However, the play analysis system 1 is applicable not only to volleyball, but also to various ball games such as soccer, baseball, table tennis, basketball, tennis, rugby, American football, lacrosse, badminton, and ice hockey. A moving object targeted by the play analysis system 1 is typically a ball. However, the moving object targeted by the play analysis system 1 is not limited to a ball, and may have a shape that does not fit the concept of "sphere", such as a badminton shuttle or an ice hockey puck.

The play analysis system 1 includes a plurality of cameras 3 (3A, 3B, 3C, 3D), a viewing terminal 20, and a play analysis device 100. FIG. Each camera 3 is connected to the play analysis device 100 through a communication network such as a wired LAN (Local Area Network) or a wireless LAN. The viewing terminal 20 is connected to the play analysis device 100 via a predetermined cable or wireless communication. Note that the play analysis device 100 may be a so-called cloud server connected via the Internet network.

As shown in FIG. 1, the cameras 3A, 3B, 3C, and 3D are installed at positions where the player's play can be photographed from the four corners of the court, from viewpoints (angles of view) looking down on the court. The number of cameras 3 may be two, three, or more than four.

Each of the cameras 3A, 3B, 3C, and 3D captures the play of the ball game and transmits the captured image 121 (see FIG. 2) to the play analysis device 100. FIG. The captured image 121 may be a frame image forming moving image data. Examples of video data formats include MPEG4, TS file, H.264 and H.264. 264 or H.264. 265.

The play analysis device 100 analyzes the captured images 121 received from the respective cameras 3A, 3B, 3C, and 3D, and presents information indicating the three-dimensional position and movement trajectory of the ball during play, and the positions and movements of the players. Play information 122 (see FIG. 2) including information and the like is generated. Based on the play information 122, the play analysis device 100 generates a play video 126 (see FIG. 2) that reproduces the movement of the ball and the movement of the players during play as a three-dimensional model.

The viewing terminal 20 displays the play video 126 output from the play analysis device 100 . Examples of the viewing terminal 20 include a PC, a tablet, VR (Virtual Reality) goggles, a dedicated viewer, and the like. As described above, since the play video 126 is composed of a three-dimensional model, the viewing terminal 20 can display the play video 126 viewed from the viewpoint specified by the user. As a result, the user can view the play of the ball game through the viewing terminal 20 from any point of view or from the selected player's point of view. A detailed description will be given below.

<Configuration of play analysis device>
FIG. 2 is a block diagram showing a configuration example of play analysis device 100 according to the present embodiment.

The play analysis device 100 includes a processor 101 , a memory 102 , a communication I/F (Interface) 103 , an input/output I/F 104 and a storage 105 .

The memory 102 stores programs and data handled by the play analysis device 100 . The memory 102 may include ROM (Read-Only Memory) and RAM (Random Access Memory).

Processor 101 realizes the functions provided by play analysis device 100 of the present embodiment by executing various processes in cooperation with other components 102-105. For example, the processor 101 implements the functions provided by the play analysis device 100 of the present embodiment by reading and executing a computer program from the memory 102 . Therefore, the processing mainly performed by the play analysis device 100 in the present embodiment may be read as the processing mainly performed by the processor 101 included in the play analysis device 100 . Also, the processor 101 may be read as other terms such as a CPU (Central Processing Unit), controller, control unit, control device, and control circuit.

A communication I/F 103 is an interface for communicating with the camera 3 . The communication I/F 103 may include at least one of a wireless communication I/F for transmitting and receiving wireless signals and a wired communication I/F to which a wired cable is connected. Wi-Fi, LTE, 4G, and 5G are examples of communication methods supported by the wireless communication I/F. Ethernet (registered trademark) is an example of a communication method supported by the wired communication I/F.

The input/output I/F 104 is an interface to which the viewing terminal 20 is connected. The input/output I/F 104 may be an interface for connecting the viewing terminal 20 with a predetermined cable, or may be an interface for connecting the viewing terminal 20 by wireless communication.

The storage 105 is an example of a non-volatile storage medium, and stores computer programs and data handled by the play analysis device 100 . Examples of the storage 105 include HDDs (Hard Disk Drives), SSDs (Solid State Drives), and flash memories.

For example, the storage 105 stores a captured image 121, play information 122, a ball model 123, a person model 124, a posture model 125, and a play video 126 as information or data. At least part of these information or data may be stored in the memory 102 .

The captured image 121 is an image captured by each camera 3 . The captured image 121 captures images of the ball, players, and the like during play. The storage 105 stores captured images 121 of each camera 3 (that is, a plurality of captured images 121).

The play information 122 includes information indicating the position of the ball during play, actions on the ball, player movements, and the like. Details of the play information 122 will be described later (see FIG. 3).

The ball model 123 is data representing a three-dimensional model of the ball. A ball model 123 may be stored in advance in the storage 105 .

The person model 124 is data representing a three-dimensional model of the player's human body. A person model 124 of each player may be stored in advance in the storage 105 .

The posture model 125 is data representing a three-dimensional model of the player's posture during play. The posture model 125 may be a three-dimensional model obtained by applying the person model 124 to the estimated skeletal posture of the player during play, as will be described later (see FIG. 4).

The play video 126 is video data that reproduces the motion of the ball and the motion of the player during play using a three-dimensional model.

<Configuration of play information>
FIG. 3 is a diagram showing a configuration example of the play information 122 according to this embodiment. FIG. 4 is a diagram for explaining the player's posture model 125. As shown in FIG.

For example, the play information 122 includes ball position information, uniform number information, team information, player position information, action information, and skeleton estimation information at each play time, as shown in FIG.

The ball position information is information indicating the three-dimensional position of the ball at the time.

Uniform number information is information indicating the uniform number of a certain player at the time. In the play information 122, uniform number information of players participating in the play may be associated with the time.

The team information is information indicating the team name of the player associated with the uniform number information.

The player position information is information indicating the three-dimensional position of the player associated with the uniform number information at the time.

The action information is information indicating the type of action performed by the player associated with the uniform number information at the time. In the case of volleyball, types of actions include serve, dig, reception, toss, attack, and block.

The skeletal structure estimation information is information obtained by estimating the posture of the skeleton during play of the player associated with the uniform number information at the relevant time. As shown in FIG. 4, the skeletal structure estimation information is composed of joint positions (white circles 127) designed based on the player's human body structure and information indicating the connection between each joint (line 128). may include information indicating the three-dimensional position of each joint of the player. Note that the skeleton estimation information may be read as skeleton information.

<Play information generation processing>
FIG. 5 is a flowchart showing an example of play information generation processing according to this embodiment.

The play analysis device 100 (processor 101) executes the following play information generation process for a plurality of captured images 121 captured at a certain time during play.

The processor 101 detects a ball from a plurality of captured images 121 (S11).

The processor 101 calculates the three-dimensional position of the ball based on the position of the ball in each captured image 121, and records it in the ball position information of the play information 122 (S12).

The processor 101 calculates the movement trajectory of the ball based on the temporal change of the three-dimensional position of the ball (S13).

The processor 101 estimates the type of player's action with respect to the ball based on the characteristics of the trajectory of the ball, and records it in the action information of the play information 122 (S14).

The processor 101 detects and tracks the player from the multiple captured images 121 (S15). Note that the processor 101 may perform the following steps S16 to S18 for each player detected and tracked in step S15.

The processor 101 identifies the three-dimensional position of the player based on the position of the player in each captured image 121, and records it in the player position information of the play information 122 (S16).

The processor 101 performs known character recognition processing on each captured image 121 to recognize the player's jersey number, and records it in the jersey number information of the play information 122 (S17).

The processor 101 performs a known skeleton estimation process on the player in each captured image 121 to estimate the posture of the player's skeleton, and records it in the skeleton estimation information of the play information 122 (S18).

The processor 101 generates the play information 122 of the entire play by performing the above processing on the captured image 121 captured at each time during the play.

It should be noted that the processing related to ball and action detection shown in steps S11 to S14 and the processing related to player detection shown in steps S15 to S18 may be executed as separate processes (for example, parallel processing). Also, the order of steps shown in FIG. 5 is an example, and the order of some steps may be changed.

<Play video generation processing>
FIG. 6 is a flowchart showing an example of play video generation processing according to the present embodiment.

The play analysis device 100 (processor 101) executes the following play video generation processing based on the play information 122 at a certain time.

The processor 101 arranges the ball model 123 at the three-dimensional position indicated by the ball position information of the play information 122 (S31).

The processor 101 identifies the person model 124 of the player corresponding to the uniform number information of the play information 122 (S32).

The processor 101 applies the specified person model 124 to the posture of the skeleton indicated by the skeleton estimation information of the play information 122 to generate a three-dimensional player posture model 125 as illustrated in FIG. 4 (S33). As a result, the processor 101 can generate the play video 126 including the posture model 125 that more realistically reproduces the physique and posture of the player.

The processor 101 arranges the generated player posture model 125 at the three-dimensional position indicated by the player position information of the play information 122 (S34).

The processor 101 generates a play video 126 including the arranged ball model 123 and the player's posture model 125 (S35).

The processor 101 performs the above processing based on the play information 122 at each time during play to generate a play video 126 of the entire play. Note that the play video 126 may use a background image at each time as a background image other than the ball and the players, or may use a representative background image acquired at a predetermined timing.

<Details of play video>
Next, the play video 126 will be described in detail.

FIG. 7 is a schematic diagram showing a display example of the play video 126. As shown in FIG. FIG. 8 is a schematic diagram showing an operation example for the play video 126. As shown in FIG.

As shown in FIG. 7, the play analysis device 100 displays a play video 126, a seek bar 130, a play button 141, a pause button 142, a slow play button 143, a rewind play button 144, and a skip forward button on the screen of the viewing terminal 20. 145, a backward skip button 146, and a track non-display button 147 are displayed.

As shown in FIG. 7, the play analysis device 100 may display a play video 126 viewed from the viewpoint of the player selected by the user (the receiver player in the case of FIG. 7). When the user selects a player displayed in the play video 126, the play analysis device 100 may switch to the play video 126 seen from the viewpoint of the selected player.

Further, the play analysis device 100 may adjust the viewing angle of the selected player according to the user's operation on the play video 126 . For example, as shown in FIG. 8, when the user performs a left swipe operation on the play video 126, the play analysis device 100 detects the play video 126 corresponding to the left viewing angle as viewed from the selected player. can be displayed. For example, as shown in FIG. 8, when the user performs a right swipe operation on the play video 126, the play analysis device 100 displays the play video 126 corresponding to the right viewing angle as viewed from the selected player. can be displayed.

Note that the swipe operation is an example of an operation when the viewing terminal 20 is compatible with a touch pad. For example, when a mouse is connected to the viewing terminal 20, the left or right swipe operation described above may be read as an operation of clicking and dragging the mouse to the left or right. Also, when the viewing terminal 20 is a VR headset, the left or right swipe operation described above may be read as an action of turning the user's face left or right.

The seek bar 130 is a bar that visually indicates the time during playback of the play video 126 . The seek bar 130 may display a marker 131 indicating the time during playback (hereinafter referred to as a playback marker). When the user moves the during-playback marker 131 , the play analysis device 100 reproduces the play video 126 from the time to which the during-playback marker 131 is moved.

In addition, the seek bar 130 may display a marker (hereinafter referred to as an action marker) 132 that indicates the time position where the action occurred. This allows the user to see the action marker 132 on the seek bar 130 and recognize at a glance the time position at which the action occurred. The color of the action marker 132 may vary depending on at least one of the player who performed the action and the type of action. Also, according to the user's (viewer's) preference, the display settings of the action markers 132 may be changed by setting the player filter button 162 or the action filter button 163, as described later. For example, the hue of action marker 132 may vary depending on the player who performed the action. For example, the brightness of action marker 132 may vary depending on the type of action. This allows the user to see the hue and brightness of the action marker 132 and grasp at a glance which player performed what kind of action at what time.

When play button 141 is touched, play analysis device 100 plays play video 126 .

When the pause button 142 is touched, the play analysis device 100 pauses the play video 126 being reproduced.

When the slow playback button 143 is touched, the play analysis device 100 performs slow playback of the play video 126 .

When the rewind play button 144 is touched, the play analysis device 100 rewinds the play video 126 for a predetermined time and plays it.

When the previous skip button 145 is touched, the play analysis device 100 places the during-playback marker 131 at the position of the action marker 132 immediately before the during-playback marker 131 (or at the position several seconds before the action marker 132). move. This allows the user to easily view the action that occurred before the play video 126 being played.

When the skip after button 146 is touched, the play analysis device 100 places the during-playback marker 131 at the position of the action marker 132 immediately after the during-playback marker 131 (or at the position several seconds before the action marker 132). move. This allows the user to easily view the action that occurred after the play video 126 being reproduced.

When track hide button 147 is pressed, play analysis device 100 hides ball track 129 in play video 126 . In this case, the trajectory non-display button 147 may be switched to a trajectory display button (not shown). When the trajectory display button is touched, play analysis device 100 displays ball trajectory 129 in play video 126 . Thereby, the user can easily switch between display and non-display of the ball trajectory 129 in the play image 126 .

FIG. 9 is a schematic diagram showing a display example of the viewpoint movement button 150. As shown in FIG.

As shown in FIG. 9 , the play analysis device 100 may display a viewpoint movement button 150 on the screen of the viewing terminal 20 . The viewpoint movement button 150 is a button for zooming the play video 126 and changing the viewpoint direction (angle). Note that the viewpoint movement button 150 may be read as a control image.

For example, the viewpoint movement buttons 150 may include a zoom in button 151 , a zoom out button 152 , a left movement button 153 , a right movement button 154 , an up button 155 and a down button 156 .

- When the zoom-in button 151 is touched, the play analysis device 100 zooms in on the play video 126 .
- When the zoom out button 152 is touched, the play analysis device 100 zooms out the play video 126 .
- When the left movement button 153 is touched, the play analysis device 100 moves the viewpoint of the play video 126 to the left.
- When the right movement button 154 is touched, the play analysis device 100 moves the viewpoint of the play video 126 to the right.
- When the raise button 155 is touched, the play analysis device 100 raises the viewpoint of the play video 126 .
- When the down button 156 is touched, the play analysis device 100 lowers the viewpoint of the play video 126 .

Thereby, the user can touch the viewpoint movement button 150 to move the viewpoint of the play video 126 to arbitrary coordinates.

Note that the user may be able to perform the same operation as the viewpoint movement button 150 by operating a keyboard, controller, or the like. For example, the play analysis device 100 performs the following operations in response to keyboard operations.

- When the w key on the keyboard is pressed, the play analysis device 100 zooms in on the play video 126 .
- When the s key on the keyboard is pressed, the play analysis device 100 zooms out the play video 126 .
- When the a key on the keyboard is pressed, the play analysis device 100 moves the viewpoint of the play video 126 to the left.
- When the d key on the keyboard is pressed, the play analysis device 100 moves the viewpoint of the play video 126 to the right.
- When the e key on the keyboard is pressed, the play analysis device 100 raises the viewpoint of the play video 126 .
- When the q key on the keyboard is pressed, the play analysis device 100 lowers the viewpoint of the play video 126 .

Thereby, the user can operate the keyboard to move the viewpoint of the play video 126 to arbitrary coordinates.

FIG. 10 is a schematic diagram showing an example of filtering action markers 132 in the seek bar 130. As shown in FIG.

As shown in FIG. 10, the play analysis device 100 may cause the viewing terminal 20 to display a player list 161 currently playing, a player filter button 162, and an action filter button 163 in response to a predetermined user operation. . As the player list 161 in play, as shown in FIG. 10, a list of players' team names, jersey numbers, and full names may be displayed.

A player filter button 162 is a button for filtering players to be displayed as action markers 132 in the seek bar 130 .

The action filter button 163 is a button for filtering the type of action displayed as the action marker 132 on the seek bar 130 .

For example, when the user selects a player from the player list 161 and touches the player filter button 162, the play analysis device 100 displays the action marker 132 corresponding to the action performed by the selected player on the seek bar 130. However, the seek bar 130 does not display the action marker 132 corresponding to the action performed by the unselected player.

For example, when the user selects a certain action type and touches the action filter button 163, the play analysis device 100 displays the action marker 132 corresponding to the selected action type on the seek bar 130, The action marker 132 corresponding to the type is not displayed on the seek bar 130.

This allows the user to display the desired action marker 132 on the seek bar 130 .

Also, the previous skip button 145 and the rear skip button 146 described above may skip to the position of the action marker 132 displayed on the seek bar 130 . Thereby, the user can easily view the play video 126 of the scene in which the desired player performed the specific action by combining the player filter button 162, the action filter button 163, the forward skip button 145, and the backward skip button 146. can do.

FIG. 11 is a schematic diagram showing an example of the player selection screen 170. As shown in FIG.

The play analysis device 100 may display a player selection screen 170 shown in FIG.

The player selection screen 170 may display a list 172 of players (for example, players registered in the list) who can participate in the play (hereinafter referred to as a player list). Also, the player selection screen 170 may display a posture model 125 of a player participating in play (for example, a player on the court).

When the user selects a plurality of players as viewpoint switching candidates from the player list 172 or posture models 125 displayed on the player selection screen 170, the play analysis device 100 displays marks corresponding to the viewpoint switching candidate players. (hereinafter referred to as viewpoint switching candidate mark) 171 may be displayed. The viewpoint switching candidate mark 171 may have a different color or pattern for each selected player. Thereby, the user can see the viewpoint switching candidate mark 171 and recognize which player it belongs to.

The play analysis device 100 may display the viewpoint switching candidate mark 171 on the screen during playback of the play video 126 shown in FIGS. 7 to 9 . Then, when the user selects the viewpoint switching candidate mark 171 , the play analysis device 100 may switch the play video 126 to the viewpoint of the player corresponding to the selected viewpoint switching candidate mark 171 . This allows the user to switch the viewpoint of the play video 126 more easily and quickly.

Note that when the user performs a predetermined operation during playback of the play video 126 shown in FIGS. 7 to 9, the play analysis device 100 temporarily stops playback of the play video 126 and displays the player selection screen 170 shown in FIG. may be displayed. Then, the user may select a player as a viewpoint switching candidate or change a player as a viewpoint switching candidate from the player selection screen 170 .

Also, the play analysis device 100 may assign a shortcut key to each of the viewpoint switching candidate marks 171 . Thereby, the user can press the shortcut key to quickly switch the viewpoint of the play video 126 .

In addition, the player selection screen 170 may display not only players participating in play (for example, players on the court), but also all players who can participate in play (for example, players registered in the roster). In this case, (A1) the viewpoint switching candidate mark 171 of a player who has not participated in the play (for example, a player who is not on the court) can be selected from the timing when the player participates in the play (for example, the timing when the player participates in the court). you can Alternatively, (A2) when the user selects the viewpoint switching candidate mark 171 of a player who has not participated in the play (for example, a player who is not on the court), the play analysis device 100 can be viewed from a waiting place (for example, a bench), or A play image 126 from a point of view from a predetermined position (for example, the position of the referee) may be displayed. The play analysis device 100 may be capable of setting which of (A1) and (A2) above is to be operated.

(Summary of this disclosure)
The content of the present disclosure can be expressed as follows.

<Expression 1>
The play analysis device 100 for analyzing sports play according to the present disclosure includes play information 122 including information on the play of a plurality of players generated based on a plurality of captured images 121 captured by at least two cameras 3. A processor 101 that reproduces the play video 126, which is a video that reproduces the play using the three-dimensional model of the player, and reproduces the play video 126 from the viewpoint indicated by the user, and the reproduced play video 126. to the viewing terminal 20 (for example, an input/output I/F 104).
This allows the user to view the sports play video 126 from any viewpoint.

<Expression 2>
In the play analysis device 100 described in expression 1, the play information 122 includes skeleton information (for example, skeleton estimation information) that is information indicating the posture of the skeleton in the play of the player, and the processor 101 calculates the skeleton of the player indicated by the skeleton information. A player's posture model 125 may be generated by applying a pre-stored three-dimensional character model 124 of the player to the posture of .
Thereby, the processor 101 can generate the play video 126 including the posture model 125 that more realistically reproduces the physique and posture of the player.

<Expression 3>
In the play analysis device 100 described in Expression 1 or 2, the processor 101 may reproduce the play video 126 with the viewpoint changed based on the user's operation on the play video 126 .
Thereby, the user can change the viewpoint of the play video 126 .

<Expression 4>
In the play analysis device 100 described in Expression 3, the processor 101 may reproduce the play video 126 from the viewpoint of the player selected by the user.
Thereby, the user can select a player and change the play video 126 seen from the viewpoint of the selected player.

<Expression 5>
In the play analysis device 100 described in expression 3, the processor 101 reproduces the play video 126 with the viewpoint changed based on the user's operation on the control image (for example, the viewpoint movement button 150) displayed on the viewing terminal 20. good.
Thereby, the user can change the viewpoint of the play video 126 by operating the control image.

<Expression 6>
In the play analysis device 100 according to any one of Expressions 1 to 5, the play information 122 includes action information that is information indicating actions of the player's play, and the processor 101 transmits the play video 126 to the viewing terminal 20. A seek bar 130 indicating the playback time may be displayed, and an action marker 132 may be displayed at the position of the time when the player selected by the user on the seek bar 130 performed an action.
Accordingly, by looking at the action marker 132 on the seek bar 130, the time position at which the action occurred can be recognized at a glance.

<Expression 7>
A play analysis method for analyzing a sports play according to the present disclosure is generated by an information processing device (for example, the play analysis device 100) based on a plurality of captured images 121 captured by at least two cameras 3. Based on the play information 122 including information on the player's play, a play video 126 that is a video reproducing the play using the three-dimensional model of the player is generated, and the play video 126 is reproduced from the viewpoint instructed by the user. and displayed on the viewing terminal 20.
This allows the user to view the sports play video 126 from any viewpoint.

<Expression 8>
A computer program for analyzing sports play according to the present disclosure includes play information 122 including information on the play of a plurality of players generated based on a plurality of captured images 121 captured by at least two cameras 3. Based on the above, a play video 126, which is a video reproducing the play using the three-dimensional model of the player, is generated, and the play video 126 from the viewpoint instructed by the user is reproduced and displayed on the viewing terminal 20. let the computer do it.
This allows the user to view the sports play video 126 from any viewpoint.

Although the embodiments have been described above with reference to the accompanying drawings, the present disclosure is not limited to such examples. It is obvious that a person skilled in the art can conceive of various modifications, modifications, substitutions, additions, deletions, and equivalents within the scope of the claims. It is understood that it belongs to the technical scope of the present disclosure. Also, the components in the above-described embodiments may be combined arbitrarily without departing from the spirit of the invention.

The technology of the present disclosure is useful for analyzing or viewing sports plays.

1 play analysis system 3, 3A, 3B, 3C, 3D camera 20 viewing terminal 100 play analysis device 101 processor 102 memory 103 communication I/F
104 input/output I/F
105 storage 121 captured image 122 play information 123 ball model 124 person model 125 posture model 126 play video 127 white circle 128 line 129 ball trajectory 130 seek bar 131 playback marker 132 action marker 141 playback button 142 pause button 143 slow playback button 144 rewind Playback button 145 Previous skip button 146 Back skip button 147 Trajectory non-display button 150 Viewpoint movement button 151 Zoom in button 152 Zoom out button 153 Left movement button 154 Right movement button 155 Up button 156 Down button 161 Player list 162 Player filter button 163 Action filter Button 170 Player selection screen 171 Viewpoint switching candidate mark 172 Player list

Claims (8)

A play analysis device for analyzing sports play,
A video reproducing the play using a three-dimensional model of the player based on play information including information on the play of the player generated based on a plurality of captured images captured by at least two cameras. a processor that generates a play video that is and reproduces the play video from a viewpoint indicated by a user;
an output interface that outputs the played play video to a viewing terminal,
Play analysis equipment. The play information includes skeleton information that is information indicating the posture of the skeleton in the play of the player,
The processor generates a posture model of the player by applying a three-dimensional human model of the player stored in advance to the posture of the skeleton of the player indicated by the skeleton information.
The play analysis device according to claim 1. The processor reproduces the play video with the viewpoint changed based on the user's operation on the play video.
The play analysis device according to claim 1 or 2. The processor reproduces the play video from the viewpoint of the player selected by the user.
The play analysis device according to claim 3. The processor reproduces the play video with the viewpoint changed based on the user's operation on the control image displayed on the viewing terminal.
The play analysis device according to claim 3 or 4. The play information includes action information that is information indicating an action of the player's play,
The processor displays a seek bar indicating the playback time of the play video on the viewing terminal, and displays an action marker at the position of the time when the player selected by the user performs an action on the seek bar.
The play analysis device according to any one of claims 1 to 5. A play analysis method for analyzing a sports play by an information processing device,
A video reproducing the play using a three-dimensional model of the player based on play information including information on the play of the player generated based on a plurality of captured images captured by at least two cameras. Generate a play video that is
playing back the play video from the viewpoint instructed by the user and displaying it on the viewing terminal;
Play analysis method. A computer program for analyzing sports play, comprising:
A video reproducing the play using a three-dimensional model of the player based on play information including information on the play of the player generated based on a plurality of captured images captured by at least two cameras. Generate a play video that is
playing back the play video from the viewpoint instructed by the user and displaying it on the viewing terminal;
A computer program that makes a computer do something.

Images