JP7001868B1 - Video analyzer, video analysis method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily grasp how the situation of victory or defeat is changing by observing the progress of a competitive game.
SOLUTION: An acquisition unit for acquiring battle video data, a conversion unit for converting battle video data into frame data of a plurality of still images, and a feature for predicting the situation of a battle game from a plurality of frame data are provided for each frame. An analysis unit that generates analysis frame data by synthesizing information indicating the judgment result of the battle game situation and the corresponding frame data, and a judgment unit that determines the situation of the battle game. A moving image analyzer equipped with an analysis moving image generator that combines a plurality of analysis frame data to generate analysis moving image data.
[Selection diagram] Fig. 2

Description

Application of Article 30, Paragraph 2 of the Patent Act (1) https: // 2424metrics. com / blog / 87 / and (2) https: // youtu. The video analyzer was released on be / EEk-KOPCFGw. (Publication date: April 18, 3rd year of Reiwa)

The present invention relates to a moving image analyzer, a moving image analysis method, and a program.

Conventionally, a competitive game in which a user and a user different from the user or a computer play against each other has been known. Such a game is configured so that the winning and losing situation changes depending on various items, events, and the like (see, for example, Patent Document 1).

Japanese Patent No. 3691832

It is often difficult for users of such games to defeat strong opponents simply by understanding the rules of the game. In addition, the user may not be able to grasp how the winning and losing situation is changing just by looking at the progress of the game, and it may be difficult to feel the joy of playing the game. there were.

Therefore, the present invention has been made in view of these points, and it is made possible to easily grasp how the situation of victory or defeat is changing by observing the progress of the competitive game. The purpose.

In the first aspect of the present invention, there is an acquisition unit that acquires a moving image displayed on the screen of a battle game as battle video data, and a conversion unit that converts the battle video data into frame data of a plurality of still images. By inputting the feature for predicting the situation of the battle game from a plurality of the frame data into the analysis model and the extraction unit for extracting the feature extracted from the frame data for each frame, the battle game A determination unit for determining the situation, an analysis frame generation unit for generating analysis frame data by synthesizing information indicating the determination result of the situation of the battle game and the corresponding frame data, and a plurality of the analysis frame data. Provided is a moving image analyzer provided with an analytical moving image generation unit that is combined to generate analytical moving image data showing a moving image obtained by analyzing the battle of the battle game.

The determination unit may determine the situation of the battle game for each frame by inputting the features extracted from each of the plurality of frame data into the analysis model.

The analysis frame generation unit converts at least a part of the image showing the decoration included in the frame data into a simple image or omits it, and the simple frame data simply showing the battle situation of the battle game. And the generated simple frame data is combined with the information indicating the determination result of the situation of the battle game as the corresponding frame data, and the analysis frame data may be generated.

The situation of the battle game determined by the determination unit may include a predicted value predicting the probability that the user who has played in the battle game will win the opponent at the time of one of the frame data.

The analysis frame generation unit determines that the situation of the battle game has changed significantly when the predicted value changes more than the threshold value in two consecutive frame data among the plurality of frame data arranged in a time series. The information shown may be combined into at least one of the two consecutive frame data.

The determination unit may further determine the technical level of the first user who has played in the battle game based on a plurality of judgment results of the situation of the battle game determined from the plurality of frame data.

The moving image analyzer has a storage unit that stores a practice drill game generated for each skill level of the user in order to improve the probability that the user wins the battle game, and the first determination unit determines. A selection unit for selecting the drill game corresponding to the technical level of one user and a notification unit for notifying the first user of information for executing the selected drill game may be further provided.

The acquisition unit acquires the moving image displayed on the screen by the first user by executing the drill game as the battle video data, and the conversion unit obtains the battle video data from the plurality of still images. Converted to frame data, the extraction unit extracts a feature for predicting the victory or defeat of the first user of the drill game from a plurality of the frame data for each frame, and the determination unit extracts the characteristics of the first user. The technical level is determined, and the selection unit selects the drill game corresponding to the technical level of the first user based on the difference between the past determination result and the current determination result of the technical level of the first user. , The notification unit may notify the first user of information for executing the selected drill game.

In the second aspect of the present invention, there is a step of acquiring a moving image displayed on the screen in a battle game executed by a computer as battle video data, and a frame of a plurality of still images of the battle video data. By inputting the step of converting into data, the step of extracting the feature for predicting the situation of the battle game from a plurality of the frame data for each frame, and the feature extracted from the frame data into the analysis model. , A step of determining the situation of the battle game, a step of synthesizing the information indicating the determination result of the situation of the battle game and the corresponding frame data to generate analysis frame data, and a plurality of the analysis frame data. Provided is a moving image analysis method including a step of generating analytical moving image data showing a moving image obtained by analyzing a battle of the battle game.

In the third aspect of the present invention, there is provided a program that, when executed by a computer, causes the computer to function as the moving image analyzer of the first aspect.

According to the present invention, by observing the progress of the competitive game, it is possible to easily grasp how the situation of victory or defeat is changing.

The outline of the moving image analyzer 100 which concerns on this embodiment is shown. A configuration example of the moving image analyzer 100 according to the present embodiment is shown. An example of the frame data converted from the battle video data by the conversion unit 132 according to the present embodiment is shown. An example of the analysis frame data generated by the analysis frame generation unit 135 according to the present embodiment is shown. An example of the analysis frame data generated by the analysis frame generation unit 135 according to the present embodiment using the simple frame data is shown. An example of the operation flow of the moving image analyzer 100 according to this embodiment is shown.

<Outline of video analyzer 100>
FIG. 1 shows an outline of the moving image analyzer 100 according to the present embodiment. The moving image analyzer 100 analyzes the moving image data of the battle game provided by the user, and supplies the user with the analyzed moving image data obtained by synthesizing the analysis result of the situation of the battle game into the moving image data. The moving image analyzer 100 is communicably connected to the user terminal 10 via the network 20.

The user terminal 10 is a terminal device owned by the user. The user terminal 10 may be a mobile terminal such as a tablet or a smartphone, and may be a personal computer, a server or the like instead. The network 20 is a communication network such as an intranet, the Internet, and a mobile phone network.

The moving image analysis device 100 is a device that analyzes the battle moving image data. The moving image analyzer 100 is, for example, a computer such as a server. The battle video data is data of a moving image displayed on the screen of the battle game by the user. The moving image analyzer 100 exchanges moving image data with the user terminal 10 via, for example, the database 30. The database 30 is a storage medium that can be accessed by the user terminal 10 and the moving image analyzer 100 via the network 20.

In the present embodiment, the user provides the battle moving image data to the moving image analysis device 100. For example, the user operates the user terminal 10 to upload the battle video data to the database 30 ((1) in FIG. 1). The battle video data is, for example, a moving image displayed on the screen of the user terminal 10 when the user operates the user terminal 10 to play a battle game against another player, a computer, or the like.

The user terminal 10 stores such a moving image as battle moving image data in an internal storage unit. As a result, the user can upload the battle video data stored in the storage unit of the user terminal 10 to the database 30. Here, it is assumed that another player can operate a player terminal communicably connected to the user terminal 10 via the network 20 to play a battle game.

Instead of this, the battle video data is a moving image displayed on a display device or the like connected to the game machine when the user operates the game machine or the like to play a battle game against another player or a computer or the like. There may be. In this case, the game machine operated by the user and the player terminal or game machine operated by another player may be communicably connected via the network 20. Further, it is desirable that the game machine operated by the user can transfer the battle video data to the storage unit of the user terminal 10 and / or the database 30, but to the user terminal 10 having a video shooting function (for example, a camera). Then, a moving image displayed on a display device or the like connected to the game machine may be captured. As a result, the user can upload the moving image displayed on the display device or the like connected to the game machine to the database 30 as the battle video data.

Then, the user operates the user terminal 10 to transmit an analysis request for the battle video data to the video analyzer 100 ((2) in FIG. 1). It is desirable that the analysis request includes the address information in which the battle video data is stored in the database 30. As a result, the video analysis device 100 can exchange a large number of relatively large-capacity video data with less labor for managing the internal storage unit, as compared with directly receiving the battle video data from the user. can. In response to such an analysis request, the moving image analyzer 100 accesses the database 30 and acquires the battle moving image data ((3) in FIG. 1).

The moving image analyzer 100 analyzes the battle moving image data and generates the analytical moving image data including the analysis result ((4) in FIG. 1). The moving image analyzer 100 may determine the technical level of the user. The moving image analyzer 100 uploads the generated analytical moving image data to the database 30 ((5) in FIG. 1). Then, the moving image analyzer 100 notifies the user terminal 10 of the address information in which the analyzed moving image data of the database 30 is stored ((6) in FIG. 1). When the moving image analyzer 100 determines the technical level of the user, the moving image analyzer 100 may notify the determination result.

As described above, the moving image analyzer 100 according to the present embodiment analyzes the result of the user fighting in the battle game, and supplies the analyzed moving image data including the analysis result to the user. As a result, the user can watch the analysis result of the video analyzer 100 based on the battle result of his / her own battle game as a moving image. The battle moving image data analyzed by the moving image analysis device 100 is not limited to the moving image actually displayed on the screen. The battle moving image data may be, for example, data output as moving image data from the user terminal 10 or a game machine, or may be moving image data calculated by simulation or the like.

In addition, in FIG. 1, an example in which the moving image analyzer 100 acquires the battle moving image data via the database 30 has been described, but the present invention is not limited to this. For example, the moving image analyzer 100 may acquire the battle moving image data from the user terminal 10 without going through the database 30. In this case, the user operates the user terminal 10 to transmit the battle moving image data from the user terminal 10 to the moving image analysis device 100. Similarly, the moving image analyzer 100 may transmit the analyzed moving image data to the user terminal 10 without going through the database 30. A more detailed configuration of the moving image analyzer 100 will be described below.

<Configuration example of video analyzer 100>
FIG. 2 shows a configuration example of the moving image analyzer 100 according to the present embodiment. The moving image analyzer 100 includes a communication unit 110, a storage unit 120, and a control unit 130.

The communication unit 110 is connected to the user terminal 10 and the database 30 via the network 20. The communication unit 110 is an interface for connecting to a communication network such as an Internet line or a mobile phone network.

The storage unit 120 is a storage medium including a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. Further, the storage unit 120 may include a large-capacity storage device such as an HDD (Hard Disk Drive) and / or an SSD (Solid State Drive). For example, when the computer functions as the moving image analyzer 100, information such as an OS (Operating System) for operating the computer and a program may be stored. Further, the storage unit 120 may store various information including a database referred to when the program is executed.

Further, the storage unit 120 may store intermediate data, calculation results, threshold values, reference values, parameters, and the like generated (or used) by the moving image analyzer 100 in the process of operation. Further, the storage unit 120 may supply the stored data to the request source in response to the request of each unit in the moving image analyzer 100. The storage unit 120 may be connected to the communication unit 110 via the network 20. For example, the storage unit 120 may include the database 30.

The storage unit 120 stores a practice drill game generated for each skill level of the user in order to improve the probability that the user wins the battle game. The storage unit 120 stores, for example, a drill game in association with a technical level. The drill game is, for example, a one-player game tool, a utility, or the like that enables a user to practice a game independently by operating an application executed on the user terminal 10. Alternatively, the drill game may be an application that can be run independently.

A drill game is, for example, a tool that reproduces a scene in which a specific item is used, a scene in which a specific event occurs, etc. in a battle game, and allows the user to experience an action that greatly affects the outcome of the game. be. In addition, the drill game is a tool that allows you to experience and learn the basic movements of the game, how to move specific items, how to link multiple items, how to generate hidden items, etc. in a battle game. You may. The drill game may be a tool that reproduces the complex aspects of a competitive game to improve the user's judgment.

The control unit 130 is, for example, a CPU (Central Processing Unit). The control unit 130 includes an acquisition unit 131, a conversion unit 132, an extraction unit 133, a determination unit 134, an analysis frame generation unit 135, an analysis moving image generation unit 136, a transmission control unit 137, a selection unit 138, and a notification unit 139. In other words, by executing the program stored in the storage unit 120, the CPU executes the acquisition unit 131, the conversion unit 132, the extraction unit 133, the determination unit 134, the analysis frame generation unit 135, the analysis video generation unit 136, and the transmission control. It functions as a unit 137, a selection unit 138, and a notification unit 139.

The acquisition unit 131 acquires the battle video data via the communication unit 110. Further, the acquisition unit 131 may acquire an analysis request for the battle video data from the user terminal 10. For example, the acquisition unit 131 controls the communication unit 110 in response to the acquisition of the analysis request, and reads and acquires the battle moving image data from the database 30. Instead of this, when the user transmits the battle video data to the video analysis device 100, the acquisition unit 131 acquires the battle video data received by the communication unit 110.

The conversion unit 132 converts the battle video data into frame data of a plurality of still images. FIG. 3 shows an example of frame data converted from the battle video data by the conversion unit 132 according to the present embodiment. Note that FIG. 3 shows an example in which the battle result of the battle game "Puyo Puyo" (registered trademark) is used as the battle video data. In FIG. 3, the area on the left side of the score of 10036 points is the area operated by the first user, and the area on the right side of the score of 13014 points is the area operated by the second user of the opponent.

For example, the score of the second user is 13014 points, which is higher than the score of the first user of 10036 points. However, due to the arrangement of items called "Puyo" that each user stacks in his / her own area, an event called "chain" occurs in which the actions that disappear when "Puyo" of the same color come into contact with each other occur continuously. The score acquired by the user differs depending on the "chain number" indicating the number of times of this "chain", and the number of occurrences of the item called "Ojamapuyo" that is generated in the opponent's area and interferes with the opponent's movement differs. ..

Therefore, it is difficult to determine the situation of the competitive game simply from the current score. In particular, even if a beginner of the game understands the rules of the game, it is possible to judge how to arrange "Puyo" and what kind of arrangement is advantageous depending on the situation. It was difficult. Therefore, the extraction unit 133 and the determination unit 134 of the present embodiment determine the user's posture from the image information shown by the frame data as shown in FIG.

The extraction unit 133 extracts features for predicting the situation of the battle game from a plurality of frame data for each frame. For example, the extraction unit 133 extracts information such as an item in the image, a user's point, an opponent's point, an event that has occurred, an item that will occur, and an event that will occur from the frame image indicated by the frame data. ..

More specifically, in the case of the frame data example shown in FIG. 3, the extraction unit 133 determines the color of the falling "puyo" and the type of the stacked "puyo" in the area of the user and the opponent. Information such as the color and arrangement, the color of "Puyo" that occurs next, the number of occurrences of "Ojamapuyo", the points of the user, and the points of the opponent are extracted.

Further, the extraction unit 133 may extract information on the number of chains and the chain score when the "puyos" are chained and disappear from the types, colors, and arrangements of the stacked "puyos". In this case, it is desirable that the extraction unit 133 extracts information on the chain score corresponding to the plurality of chains. For example, the extraction unit 133 extracts information on the number of chains and the chain score of three sets in descending order of the number of chains from the areas of the user and the opponent, respectively. It is desirable that the extraction unit 133 quantifies the extracted features into feature quantities. Further, the extraction unit 133 may normalize the quantified feature amount.

The determination unit 134 determines the situation of the battle game by inputting the features extracted from the frame data into the analysis model. The analysis model is a learning model in which, for example, the feature amount of the feature extracted by logistic regression analysis is input as an explanatory variable, and the probability that the user wins the opponent is predicted as the objective variable. The analysis model is, for example, SVM (Support Vector Machine).

It is desirable that the analytical model has been trained using the features extracted from the training data so as to improve the prediction accuracy of the user's winning probability. The learning data is, for example, battle video data when a skilled player plays a battle game, battle video data when a player who can win more opponents plays a battle game, and a player outputs a high score. It is the battle video data of the time. Further, the analysis model may be a model for calculating the winning probability of the user by using a predetermined algorithm.

The determination unit 134 determines the situation of the battle game for each frame, for example, by inputting the features extracted from each of the plurality of frame data into the trained analysis model. As a result, the situation of the battle game determined by the determination unit 134 includes a predicted value that predicts the probability that the user who has played in the battle game will win the opponent at the time of one frame data. Such a predicted value can be improved in accuracy by learning a learning model, and since it is an objective evaluation, it is suitable to be used as an index showing the situation of a competitive game.

The determination unit 134 may further determine the technical level of the user who has played in the battle game based on a plurality of judgment results of the situation of the battle game determined from the plurality of frame data. The determination unit 134 uses, for example, the average value of the user's posture from the start to the end of the battle game as the user's technical level. Further, the determination unit 134 may determine that the skill level of the user is lower as the time during which the user's posture becomes lower than the threshold value is faster after the start of the battle game.

Further, the determination unit 134 may determine the user's skill level, operation speed, etc. by comparing the threshold value with the number of times the user presses the operation button, the speed, and the like. The determination unit 134 may determine the event generation ability of the user based on the number of occurrences of events such as the number of chains. The determination unit 134 may include, for example, the result of comparison with an ideal player in the determination result.

The analysis frame generation unit 135 generates analysis frame data by synthesizing the information indicating the determination result of the situation of the battle game and the corresponding frame data. The analysis frame generation unit 135, for example, synthesizes one analysis frame data by synthesizing the situation of the battle game determined by the determination unit 134 using the features extracted from the one frame data into the one frame data. Generate.

<Example of analysis frame data>
FIG. 4 shows an example of analysis frame data generated by the analysis frame generation unit 135 according to the present embodiment. At the time of the analysis frame data shown in FIG. 4, the probability that the user operating the area on the left side wins is 60%, and it can be seen that the development is advantageous to the user. Note that FIG. 4 shows an example in which the analysis frame generation unit 135 synthesizes an image showing the situation of the battle game using numbers and a bar graph into the frame data, but the present invention is not limited to this. The analysis frame generation unit 135 may represent the situation of the battle game by using numbers, pictograms, or graphs. Further, the analysis frame generation unit 135 may be expressed by using a pie chart, a line graph, or the like.

Here, the analysis frame generation unit 135 may process the corresponding frame data and synthesize the processed frame data and the situation of the battle game. The analysis frame generation unit 135, for example, converts at least a part of the image showing the decoration contained in the frame data into a simple image or omits it, and the simple frame data showing the battle situation of the battle game simply. To generate. Then, the analysis frame generation unit 135 synthesizes the generated simple frame data as the corresponding frame data with the information indicating the determination result of the situation of the battle game, and generates the analysis frame data.

<Example of analysis frame data using simple frame data>
FIG. 5 shows an example of analysis frame data generated by the analysis frame generation unit 135 according to the present embodiment using the simple frame data. It can be seen that in the analysis frame data shown in FIG. 5, the background image is deleted and the background image is simplified as compared with the analysis frame data shown in FIG.

The analysis frame generation unit 135 can reduce the capacity of the analysis frame data by using such simple frame data. Further, since the analysis frame generation unit 135 generates the analysis frame data without using a part of the image shown by the battle video data, the right holder or the like needs permission to use the part of the image. Even in this case, analysis frame data can be generated and used without permission.

The analysis video generation unit 136 combines a plurality of analysis frame data to generate analysis video data showing a moving image obtained by analyzing the battle of the battle game. Since the analysis frame data is combined with the analysis frame data in which the determination result of the posture determined for each frame is combined, the analysis video data is generated so that the user's posture changes at the same time as the video progresses.

The transmission control unit 137 controls the communication unit 110 to upload the analysis video data generated by the analysis video generation unit 136 to the database 30. Further, the transmission control unit 137 may transmit the analysis moving image data to the user terminal 10. When the user designates the storage area of the analysis moving image data, the transmission control unit 137 may transmit and store the analysis moving image data to the storage area designated by the user.

The selection unit 138 selects a drill game corresponding to the skill level of the user determined by the determination unit 134. The selection unit 138 reads, for example, a drill game corresponding to the determination result of the user's technical level from the storage unit 120 and uses it as the selection result. Here, the transmission control unit 137 may control the communication unit 110 to upload the drill game selected by the selection unit 138 to the database 30, and instead, transmit the drill game to the user terminal 10. May be good. Further, the transmission control unit 137 may transmit and store the drill game in the storage area designated by the user.

The notification unit 139 notifies the user that the analysis video data obtained by analyzing the battle video data has been generated. In addition, the notification unit 139 notifies the user of information for executing the selected drill game. The notification unit 139, for example, controls the communication unit 110 to transmit a message including the determination result of the user's technical level and the recommendation to execute the corresponding drill game to the user terminal 10. Further, the notification unit 139 may notify the information of the address where the drill game is stored, the transmission of the drill game, and the like.

The moving image analysis device 100 according to the above embodiment can generate analysis moving image data obtained by analyzing the battle moving image data provided by the user and provide the analysis moving image data to the user. As a result, the user can easily grasp how the situation of victory or defeat is changing by observing the progress of the competitive game that he / she has operated.

In addition, the moving image analyzer 100 may display in an easy-to-understand manner to the user that the situation of victory or defeat has changed significantly. For example, the determination unit 134 further determines that the predicted value of the probability that the user will win the opponent has changed more than the threshold value in two consecutive frame data among the plurality of frame data arranged in the time series. Then, the analysis frame generation unit 135 outputs information indicating that the situation of the competitive game has changed significantly to at least one of the two consecutive frame data when the predicted value changes more than the threshold value in the two consecutive frame data. Synthesize.

The analysis frame generation unit 135 changes, for example, the display color of the situation of the battle game. Further, the analysis frame generation unit 135 may significantly change the display area of the situation of the battle game. The analysis frame generation unit 135 may synthesize a message, a voice, or the like indicating that the situation of the battle game has changed significantly. Further, the analysis frame generation unit 135 may duplicate the frame data in which the situation of the battle game has changed significantly and include it in the analysis frame data. As a result, the moving image analysis device 100 allows the user to confirm the scene for determining the victory or defeat of the competitive game at a slow speed.

From these, the user can understand the points that determine the victory or defeat of the competitive game. Therefore, even if the user fights against a strong opponent and loses, the moving image analyzer 100 can easily generate positive emotions that make watching the operation of an advanced person enjoyable.

Further, since the video analysis device 100 notifies the technical level of the user, the user can quantitatively grasp the proficiency level of the competitive game. Then, since the video analyzer 100 recommends a drill game corresponding to the user's technical level, what the user should do in order to grasp the knack of the game, acquire the technique, improve the technique, and the like efficiently. It's easy to understand what should be done. In this way, the moving image analyzer 100 can facilitate the improvement of the user's technique.

<Example of operation flow of video analyzer 100>
FIG. 6 shows an example of the operation flow of the moving image analyzer 100 according to the present embodiment. The operation flow shown in FIG. 6 is an example of an operation flow started by the user uploading the battle moving image data to the database 30 and transmitting a data analysis request to the moving image analysis device 100.

First, the acquisition unit 131 reads the battle video data from the database 30 and acquires it (S51). Next, the conversion unit 132 converts the battle video data into frame data of a plurality of still images (S52). Next, the extraction unit 133 extracts a feature for predicting the situation of the battle game from a plurality of frame data for each frame (S53).

Next, the determination unit 134 determines the situation of the battle game by inputting the features extracted from the frame data into the analysis model (S54). Further, the determination unit 134 further determines the technical level of the user who has played in the battle game based on the plurality of judgment results of the situation of the battle game determined from the plurality of frame data (S55).

Next, the analysis frame generation unit 135 generates analysis frame data by synthesizing the information indicating the determination result of the situation of the battle game and the corresponding frame data (S56). Next, the analysis video generation unit 136 combines a plurality of analysis frame data to generate analysis video data showing a moving image obtained by analyzing the battle of the battle game (S57). Next, the selection unit 138 selects a drill game corresponding to the technical level of the user determined by the determination unit 134 (S58).

Next, the transmission control unit 137 uploads the analysis video data generated by the analysis video generation unit 136 to the database 30 (S59). Here, the transmission control unit 137 uploads the drill game selected by the selection unit 138 to the database 30. Next, the notification unit 139 notifies the user that the analysis video data obtained by analyzing the battle video data has been generated (S60). Here, the notification unit 139 notifies the user of information for executing the selected drill game.

As a result, the user can download the analysis video data corresponding to the battle video data and confirm the analysis result by the video analyzer 100. The user can also download the drill game and run the drill game. The moving image analyzer 100 may be stopped in operation if there is no further analysis request from the user (S61: Yes).

When the moving image analyzer 100 obtains a further analysis request from the user (S61: No), the moving image analyzer 100 returns to S31 and starts a flow of generating the analysis moving image data based on the next battle moving image data. Here, the user may upload the result of executing the drill game to the database 30 as the next analysis moving image data.

<Example of operation flow using the result of executing a drill game as battle video data>
In this case, in S51, the acquisition unit 131 acquires the moving image displayed on the screen as the battle moving image data by the user executing the drill game. Next, in S52, the conversion unit 132 converts the battle moving image data into frame data of a plurality of still images. Next, in S53, the extraction unit 133 extracts a feature for predicting the victory or defeat of the user of the drill game from a plurality of frame data for each frame.

Next, the determination unit 134 determines the situation of the battle game in S54, and determines the skill level of the user in S55. Next, in S56, the analysis frame generation unit 135 generates analysis frame data. Next, in S57, the analysis moving image generation unit 136 generates the analysis moving image data.

Next, in S58, the selection unit 138 selects a drill game corresponding to the user's technical level based on the difference between the past determination result and the current determination result of the user's technical level. Next, in S59, the transmission control unit 137 uploads the analysis moving image data and the drill game to the database 30. Next, in S60, the notification unit 139 notifies the user of the fact that the analysis moving image data has been generated and the information for executing the selected drill game.

As described above, the moving image analyzer 100 according to the present embodiment may generate analysis moving image data based on the result of the user executing the drill game and supply it to the user. As a result, the user can repeat the execution of the drill game and the confirmation of the corresponding analysis video data, and the ability of the game can be improved more efficiently. In addition, the user can efficiently enjoy the actual battle by executing the battle game and confirming the corresponding analysis video data in combination with the execution of the drill game and the corresponding analysis video data. You can improve the ability of the game.

Although the moving image analyzer 100 has described an example of recommending a drill game corresponding to the technical level of the user, the present invention is not limited to this. The moving image analyzer 100 may recommend exemplary image data corresponding to the technical level of the user. The exemplary image data is moving image data including a scene in which an advanced player in a competitive game effectively generates a specific event, a scene in which a specific item is effectively operated, and the like.

Before a user becomes an advanced player in a competitive game, it may be necessary to improve various techniques. Therefore, it is conceivable to prepare a plurality of moving images in advance to introduce various technologies and have the user view them. However, watching a lot of videos to learn a lot of techniques takes time, and the user may not enjoy it because he / she does not actually execute the game. Therefore, the moving image analyzer 100 recommends model image data corresponding to the technical level of the user, so that the user can see the appropriate technology corresponding to the technical level at an appropriate timing, and the technology is improved. Can be advanced efficiently.

Further, the drill game may be a tool that reproduces the aspect of the game indicated by such exemplary image data. As a result, the user can practice in the same phase as the aspect of the game properly operated by the advanced player, so that the user can learn the operation method of the advanced player.

In the moving image analysis device 100 according to the above embodiment, the determination unit 134 has described an example of determining the situation of the battle game for each frame, but the present invention is not limited to this. The determination unit 134 may determine the position of the plurality of frames based on the characteristics of the plurality of frames. For example, the determination unit 134 may calculate the situation of a plurality of frames by inputting the features of the plurality of frames into the analysis model. Further, the determination unit 134 may calculate the average value or the like of the plurality of positions determined for each frame to determine the positions of the plurality of frames.

Further, the determination unit 134 does not have to determine the situation of the battle game for all the frames. For example, the determination unit 134 extracts one frame for each set of a plurality of n frames arranged in a time series, and determines the situation of the battle game with respect to the extracted frames. Here, n is an integer of 2 or more. The determination unit 134 may use the determination result of the shape of the extracted frame as the determination result of the shape of the n-1 frames that have not been extracted.

In this case, the analysis frame generation unit 135 may generate the analysis frame data using all the frames, or may generate the analysis frame data using only the extracted frames. When the analysis frame data is generated using only the frames extracted by the analysis frame generation unit 135, the analysis video generation unit 136 can generate the analysis video data having a capacity smaller than the capacity of the battle video data.

In the video analysis device 100 according to the above embodiment, "Puyo Puyo" has been described as an example of a battle game, but the present invention is not limited to this. The battle game may be a game in which the state of the battle is displayed on the display screen by manipulating items, characters, etc. as the game progresses to generate items, events, etc. while competing against the opponent. By acquiring the image data displayed on the display screen in such a battle game, the moving image analyzer 100 can determine the situation of the battle game and output it as analysis image data.

Although the present invention has been described above using the embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments, and various modifications and changes can be made within the scope of the gist. be. For example, all or part of the device can be functionally or physically distributed / integrated in any unit. Also included in the embodiments of the present invention are new embodiments resulting from any combination of the plurality of embodiments. The effect of the new embodiment produced by the combination has the effect of the original embodiment together.

10 User terminal 20 Network 30 Database 100 Video analyzer 110 Communication unit 120 Storage unit 130 Control unit 131 Acquisition unit 132 Conversion unit 133 Extraction unit 134 Judgment unit 135 Analysis frame generation unit 136 Analysis video generation unit 137 Transmission control unit 138 Selection unit 139 Notification section

Claims (8)

An acquisition unit that acquires the moving image displayed on the screen of the battle game as battle video data,
A conversion unit that converts the battle video data into frame data of a plurality of still images, and
An extraction unit that extracts features for predicting the situation of the battle game from a plurality of the frame data for each frame, and an extraction unit.
By inputting the features extracted from the frame data into the analysis model, the situation of the battle game is determined , and further, based on a plurality of determination results of the situation of the battle game determined from the plurality of frame data. A determination unit that determines the technical level of the first user who has played in the battle game ,
An analysis frame generation unit that generates analysis frame data by synthesizing information indicating a determination result of the situation of the battle game and the corresponding frame data.
With an analysis video generation unit that combines a plurality of the analysis frame data to generate analysis video data showing a moving image that analyzes the battle of the battle game.
A storage unit that stores a practice drill game generated for each user's skill level in order to improve the probability that the user will win the battle game.
A selection unit that selects the drill game corresponding to the technical level of the first user determined by the determination unit, and a selection unit.
With a notification unit that notifies the first user of information for executing the selected drill game.
A video analyzer equipped with. The determination unit determines the situation of the battle game for each frame by inputting the features extracted from each of the plurality of frame data into the analysis model.
The moving image analyzer according to claim 1. The analysis frame generator is
At least a part of the image showing the decoration contained in the frame data is converted into a simple image or omitted to generate simple frame data that simply shows the battle situation of the battle game.
The generated simple frame data is combined with the information indicating the determination result of the situation of the battle game as the corresponding frame data, and the analysis frame data is generated.
The moving image analyzer according to claim 1 or 2. The situation of the battle game determined by the determination unit includes a predicted value predicting the probability that a user who has played in the battle game will win the opponent at the time of one of the frame data, according to claims 1 to 3. The moving image analyzer according to any one of the items. The analysis frame generation unit determines that the situation of the battle game has changed significantly when the predicted value changes more than the threshold value in two consecutive frame data among the plurality of frame data arranged in a time series. Combining the information shown into at least one of two consecutive frame data,
The moving image analyzer according to claim 4. The acquisition unit acquires the moving image displayed on the screen by the first user executing the drill game as the battle video data.
The conversion unit converts the battle video data into the frame data of a plurality of still images.
The extraction unit extracts, for each frame, features for predicting the victory or defeat of the first user of the drill game from the plurality of frame data.
The determination unit determines the technical level of the first user, and determines the technical level of the first user.
The selection unit selects the drill game corresponding to the technical level of the first user based on the difference between the past determination result and the current determination result of the technical level of the first user.
The notification unit notifies the first user of information for executing the selected drill game.
The moving image analyzer according to any one of claims 1 to 5 . Computer runs,
The step of competing in a battle game and acquiring the moving image displayed on the screen as battle video data,
The step of converting the battle video data into frame data of a plurality of still images,
A step of extracting features for predicting the situation of the battle game from a plurality of the frame data for each frame, and
A step of determining the situation of the battle game by inputting the feature extracted from the frame data into the analysis model, and
A step of further determining the technical level of the first user who has played in the battle game based on a plurality of judgment results of the situation of the battle game determined from the plurality of frame data.
A step of synthesizing the information indicating the determination result of the situation of the battle game and the corresponding frame data to generate the analysis frame data.
A step of combining a plurality of the analysis frame data to generate analysis video data showing a moving image obtained by analyzing the battle of the battle game.
A step of selecting the drill game corresponding to the technical level of the first user from a storage unit that stores a drill game for practice generated for each technical level of the user, and a step of selecting the drill game.
A step of notifying the first user of information for executing the selected drill game.
A video analysis method. A program that, when executed by a computer, causes the computer to function as the moving image analyzer according to any one of claims 1 to 6 .

Images