Apparatus and method for detecting anomaly in sports match, and recording medium on which instructions are stored

Abstract

According to at least one embodiment of the present disclosure, an apparatus for detecting whether an anomaly has occurred in a sports match can be presented. The apparatus according to the present disclosure comprises: a communication circuit; one or more processors; and one or more memories in which instructions to be executed by the one or more processors are stored, wherein, when the instructions are executed, the one or more processors: acquire movement data related to a movement of an object in a target sports match and / or match situation data indicating a match situation of the target sports match; detect, on the basis of the movement data and / or the match situation data, whether an anomaly has occurred in the target sports match; and acquire analysis data indicating whether the anomaly has occurred.

Description

Recording medium recording a device, method and command for detecting anomalies in sports games

[0001] The present disclosure relates to a technique for detecting anomalies in sports games.

[0002] As the popularity of sports continues to grow, the size of the sports industry is also expanding. However, capitalizing on this popularity, instances of unfair advantage gained through unfair practices such as match fixing have emerged, undermining the fairness of sports competitions. Cheating in sports goes beyond simply disrupting the enjoyment of the sport; it undermines the efforts of the players, coaches, and managers involved, undermining the spirit of the sport.

[0003] Meanwhile, sports matches are constantly subject to changes in player behavior patterns and the flow of the game due to various variables, such as player injuries and tactical changes. Furthermore, cheating can occur even in situations that don't directly impact the outcome of the match.

[0004] At least one embodiment of the present disclosure provides a technique for detecting the presence or absence of an abnormality in a sports game based on various data.

[0005] In one aspect of the present disclosure, a device for detecting an abnormality in a sports game may be proposed. The device according to the present disclosure may include a communication circuit; one or more processors; and one or more memories storing instructions to be executed by the one or more processors, wherein, upon execution of the instructions, the one or more processors may be configured to obtain at least one of movement data regarding the movement of an object in a target sports game or game situation data indicating a game situation of the target sports game, and, based on at least one of the movement data or the game situation data, detect an abnormality in the target sports game and obtain analysis data indicating the abnormality. In addition, the one or more processors may be configured to receive image data regarding the target sports game from a camera group including one or more cameras through the communication circuit and generate the movement data based on the image data when obtaining the movement data.

[0006] In one embodiment, the one or more processors may be configured to receive, from another device via the communication circuit, the match situation data indicating the match situation of the target sporting event, in obtaining the match situation data.

[0007] In one embodiment, the one or more processors may be configured to receive image data from the camera group through the communication circuit and generate the game situation data based on the image data in obtaining the game situation data.

[0008] In one embodiment, the one or more processors may be configured to transmit at least one of the movement data, the match situation data, or the analysis data to a user terminal.

[0009] In one embodiment, the one or more processors may be configured to input at least one of the movement data or the game situation data into a learning model trained to detect the presence or absence of an abnormality in a sports game, and obtain the analysis data indicating the presence or absence of an abnormality in the target sports game as an output of the learning model.

[0010] In one embodiment, the learning model may be a model trained using the first movement pattern of the object in one or more other sports events as learning data. Furthermore, the one or more processors may be configured to extract a second movement pattern of the object in the target sports event based on the movement data, and obtain the analysis data from the second movement pattern using the learning model.

[0011] In one embodiment, the learning model may be a model trained using a first match situation pattern of another sporting event as training data. Furthermore, the one or more processors may be configured to extract a second match situation pattern of the target sporting event based on the match situation data, and to use the learning model to obtain the analysis data from the second match situation pattern.

[0012] In one embodiment, the learning model may be a model trained using first match situation patterns from each of a plurality of different sports games as learning data. Furthermore, the one or more processors may be configured to extract second match situation patterns of the target sports game based on the match situation data, and to obtain analysis data from the second match situation patterns using the learning model.

[0013] In one embodiment, the image data may include an image set divided into frames by capturing images of the target sporting event from one or more angles by the camera group.

[0014] In one embodiment, the one or more processors may be configured to generate frame-by-frame position data of the object of the target sporting event based on the image set. Furthermore, the frame-by-frame position data may include three-dimensional coordinates of the object's location.

[0015] In one embodiment, the one or more processors may be configured to generate movement data of the object based on the frame-by-frame position data. In addition, the movement data may include at least one of a movement direction, a movement speed, a movement time, or a movement trajectory of the object.

[0016] In one embodiment, the object may include at least one of a player or a ball of the target sporting event.

[0017] In one embodiment, the match situation data may include at least one of the match time, score, number of rule violations, match lead time, or referee decision result in the target sport match.

[0018] In one aspect of the present disclosure, a method for detecting an abnormality in a sports game may be proposed. The method according to the present disclosure may be a method performed by a device including one or more processors and one or more memories storing instructions to be executed by the one or more processors. The method according to the present disclosure may include a step in which the one or more processors acquire at least one of movement data regarding the movement of an object in a target sports game or game situation data indicating a game situation of the target sports game; and a step in which the one or more processors detect an abnormality in the target sports game and acquire analysis data indicating the abnormality based on at least one of the movement data or the game situation data. In addition, the step of acquiring the movement data may include a step in which the one or more processors receive image data regarding the target sports game from a camera group including one or more cameras; and a step in which the movement data is generated based on the image data.

[0019] In one aspect of the present disclosure, a non-transitory computer-readable recording medium having recorded thereon commands for detecting an abnormality in a sports game may be proposed. The commands recorded in the non-transitory computer-readable recording medium according to the present disclosure are commands to be executed by one or more processors, and when executed by the one or more processors, the one or more processors may obtain at least one of movement data regarding the movement of an object in a target sports game or game situation data indicating a game situation of the target sports game, and detect an abnormality in the target sports game based on at least one of the movement data or the game situation data, and obtain analysis data indicating the abnormality. In addition, when obtaining the movement data, the commands may cause the one or more processors to receive image data regarding the target sports game from a camera group including one or more cameras, and generate the movement data based on the image data.

[0020] According to at least one embodiment of the present disclosure, it is possible to detect the presence or absence of an abnormality in a sports game based on various data.

[0021] The effects according to the technical idea of ​​the present disclosure are not limited to the effects mentioned above, and other effects not mentioned can be clearly understood by a person skilled in the art from the description of the specification.

[0022] Figure 1 is a drawing showing the operation process of a device according to one embodiment of the present disclosure.

[0023] FIG. 2 is a block diagram of a device according to one embodiment of the present disclosure.

[0024] FIG. 3 is a diagram illustrating a process for generating position data of an object according to one embodiment of the present disclosure.

[0025] FIG. 4 is a diagram illustrating a process for generating movement data of an object according to one embodiment of the present disclosure.

[0026] FIG. 5 is a diagram illustrating a process for generating movement data of an object according to one embodiment of the present disclosure.

[0027] FIG. 6 is a diagram illustrating a process for generating position data of an object according to one embodiment of the present disclosure.

[0028] FIG. 7 is a diagram illustrating a process for generating movement data of an object according to one embodiment of the present disclosure.

[0029] FIG. 8 is a diagram illustrating a process of obtaining analysis data using a learning model learned to detect the presence or absence of an abnormality in a sports game according to one embodiment of the present disclosure.

[0030] FIG. 9 is a diagram illustrating a process for obtaining analysis data according to one embodiment of the present disclosure.

[0031] FIG. 10 is a diagram illustrating a process for obtaining analysis data according to one embodiment of the present disclosure.

[0032] FIG. 11 is a diagram illustrating a process for obtaining analysis data according to one embodiment of the present disclosure.

[0033] FIG. 12 is a diagram illustrating a method for detecting the presence or absence of an abnormality in a sports game according to one embodiment of the present disclosure.

[0034] The various embodiments described in this disclosure are exemplified for the purpose of clearly explaining the technical concept of this disclosure and are not intended to be limited to specific embodiments. The technical concept of this disclosure includes various modifications, equivalents, alternatives, and embodiments selectively combined from all or part of the embodiments described in this disclosure. Furthermore, the scope of the technical concept of this disclosure is not limited to the various embodiments presented below or the specific descriptions thereof.

[0035] Terms used in this disclosure, including technical or scientific terms, unless otherwise defined, may have the meaning commonly understood by a person of ordinary skill in the art to which this disclosure belongs.

[0036] In this disclosure, expressions such as "includes," "may include," "comprises," "may have," "has," and "may have" indicate the presence of a target feature (e.g., a function, operation, or component), but do not exclude the presence of other additional features. In other words, such expressions should be understood as open-ended terms that imply the possibility of including other embodiments.

[0037] In this disclosure, singular expressions may include plural meanings unless the context clearly indicates otherwise, and this also applies to singular expressions described in the claims.

[0038] In this disclosure, expressions such as “first,” “second,” or “first,” “second,” etc., unless the context indicates otherwise, are used to distinguish one object from another when referring to multiple similar objects, and do not limit the order or importance among the objects.

[0039] In this disclosure, expressions such as “A, B, and C,” “A, B, or C,” “A, B, and / or C,” or “at least one of A, B, and C,” “at least one of A, B, or C,” “at least one of A, B, and / or C,” “at least one selected from A, B, and C,” “at least one selected from A, B, or C,” “at least one selected from A, B, and / or C,” etc., can refer to each listed item or all possible combinations of the listed items. For example, “at least one selected from A and B” can refer to (1) A, (2) at least one of A, (3) B, (4) at least one of B, (5) at least one of A and at least one of B, (6) at least one of A and B, (7) at least one of B and A, and (8) both A and B.

[0040] In this disclosure, the expression “based on or according to” is used to describe one or more factors that influence a decision, act of judgment, or action described in a phrase or sentence containing the expression, and the expression does not exclude additional factors that influence the decision, act of judgment, or action.

[0041] In the present disclosure, determining C information based on A information and B information means taking A information and B information into consideration in determining C information, and does not exclude that information other than A information and B information is additionally considered.

[0042] In the present disclosure, the expression that a component (e.g., a first component) is “connected” or “connected” to another component (e.g., a second component) may mean that the component is directly connected or connected to the other component, as well as connected or connected via a new other component (e.g., a third component).

[0043] In the present disclosure, “configured to” may have the meaning of “set to”, “having the ability to”, “modified to”, “made to”, “capable of”, etc., depending on the context. The expression is not limited to the meaning of “specifically designed in hardware”, and for example, a processor configured to perform a specific operation may mean a special purpose computer structured through programming to perform the specific operation.

[0044] In the present disclosure, a "learning model" may be designed to implement the structure of a human brain on a computer, and may include a plurality of network nodes that simulate neurons of a human neural network and have weights. The plurality of network nodes simulate the synaptic activity of neurons that exchange signals through synapses and may have connections among themselves. In the learning model, the plurality of network nodes may be located at layers of different depths and exchange data according to convolutional connections. For example, the learning model may be an artificial neural network model, a time series analysis model, etc. Here, the time series analysis model is a model that analyzes time series data, and may be a recurrent neural network (RNN) model, a long short-term memory (LSTM) model, or an Autoregressive Integrated Moving Average (ARIMA) model. Meanwhile, the present disclosure is not limited thereto, and various types of models for analyzing data may be used.

[0045] In the present disclosure, a “training process” may mean a process in which a learning model extracts and analyzes features (patterns) of input data and output data pairs of learning data, repeats the process of deriving correlations between input and output data, and optimizes parameters of the learning model based on the correlations between input and output data.

[0046] In the present disclosure, an “inference process” may mean a process in which a learning model applies a previously learned pattern to new input data to generate output data as a result of prediction or classification of the input data.

[0047] Hereinafter, various embodiments of the present disclosure will be described with reference to the attached drawings. In the attached drawings and the description of the drawings, identical or substantially equivalent components may be assigned the same reference numerals. Furthermore, in the description of various embodiments below, duplicate descriptions of identical or corresponding components may be omitted, but this does not mean that the corresponding components are not included in the embodiments.

[0048] FIG. 1 is a diagram illustrating an operation process of a device (100) according to one embodiment of the present disclosure. In the present disclosure, the device (100) may be a device (e.g., a server) that detects anomalies in a sports game.

[0049] In the present disclosure, a camera group (130) may include one or more cameras (131, 132, 133, 134, 135, 136). The camera group (130) may capture a game space (110) of a sports game from one or more angles through one or more cameras (131, 132, 133, 134, 135, 136), thereby generating image data (160) of the corresponding sports game.

[0050] For example, the game space (110) may refer to a predetermined three-dimensional space related to a sports game. Specifically, the game space (110) may include an area where a sports game takes place (e.g., a court), an area where objects (120) of the sports game may be located, an area where spectators, analysts, or referees of the sports game may be located, an area where stadium facilities (e.g., an electronic scoreboard) may be located, etc. Meanwhile, the object (120) may include, for example, at least one of a player or a ball of the sports game.

[0051] For example, at least some of the cameras (131, 132, 133, 134, 135, 136) of the camera group (130) may be fixed cameras, and the others may be tracking cameras. The fixed camera may refer to a camera that photographs at least a portion of the playing space (110) at a predetermined angle, magnification, etc. The tracking camera may be a camera that tracks and photographs at least one object (120) within the playing space (110). The tracking camera may track and photograph the object (120) while moving according to a preset, and the preset may be one or more settings for camera parameters such as a shooting angle, magnification, etc. Meanwhile, FIG. 1 illustrates six cameras (131, 132, 133, 134, 135, 136) that photograph the playing space (110), but the present disclosure is not limited thereto. The number and arrangement of cameras within a camera group (130) may change depending on the type of sport, the shape of the stadium (e.g., size), the number of spectators, etc.

[0052] For example, each camera of the camera group (130) can capture a game space (110) of a sports game at a predetermined frame rate (e.g., 120 fps) to generate image data (160) of the corresponding sports game. Here, the image data (160) can include an image set obtained by dividing the captured video of the sports game into units of shooting frames (e.g., 120 fps).

[0053] In the present disclosure, the external device (140) may be a device that stores or manages game situation data (170) indicating the game situation of a sports game. For example, the external device (140) may be a database, a server that controls stadium facilities (e.g., an electronic scoreboard), etc.

[0054] In the present disclosure, the external device (150) may be a terminal of a user (e.g., analyst, viewer) using a sports game broadcasting service.

[0055] In one embodiment, the device (100) can detect the presence or absence of an anomaly in a target sporting event and generate analysis data (190) indicating the presence or absence of an anomaly. Here, the target sporting event may refer to one or more sporting events for which an anomaly is to be detected. To this end, the device (100) can perform at least some of the following operations.

[0056] For example, the device (100) can receive image data (160) for a target sporting event from a camera group (130).

[0057] For example, the device (100) can generate frame-by-frame position data of an object (120) in a target sports game based on image data (160). Here, the frame-by-frame position data can include three-dimensional coordinates where the object (120) is located.

[0058] For example, the device (100) may generate movement data regarding the movement of an object (120) in a target sports game based on frame-by-frame position data. Here, the movement data of the object (120) may include metrics indicating the movement of the object (120), such as the direction of movement, movement speed, movement time, and movement trajectory of the object (120) in the target sports game.

[0059] For example, the device (100) may receive game situation data (170) indicating the game situation of a target sports game from an external device (140). Here, the game situation data (170) may include indicators indicating the game situation of the target sports game, such as the game progress time, score, number of rule violations, game leading time, and referee decision results in the target sports game.

[0060] Additionally or alternatively, the device (100) may generate game situation data (170) based on image data (160) received from the camera group (130). For example, the device (100) may extract an image of a scoreboard of a target sporting event from an image set included in the image data (160), and identify indicators indicating the game situation of the target sport, such as game time, score, number of rule violations, game leading time, and referee decision results, from the extracted image to generate game situation data (170).

[0061] For example, the device (100) can detect the presence or absence of an abnormality in a target sports game based on at least one of movement data (180) of an object (120) in the target sports game or game situation data (170) of the target sports game, and generate analysis data (190) indicating the presence or absence of an abnormality.

[0062] For example, the device (100) can transmit at least one of movement data (180), game situation data (170), or analysis data (190) to an external device (150). Through this, a user can check the movement data (180), game situation data (170), analysis data (190), etc. displayed on the display of the external device (150) and determine whether an abnormality is detected in the target sports game.

[0063] Sports matches are constantly changing due to various variables, such as player injuries and tactical changes, which can make it difficult to detect cheating simply by watching the game. Furthermore, cheating can occur even in situations that don't directly impact the outcome of the match, making it extremely difficult to detect such cheating directly without third-party reports.

[0064] According to the technology for detecting the presence or absence of an abnormality in a sports game of the present disclosure, the device (100) analyzes at least one of image data of the sports game or game situation data of the sports game that occurred while the sports game was in progress, thereby obtaining analysis data indicating the presence or absence of an abnormality in the sports game, thereby effectively and accurately detecting the presence or absence of an abnormality in the sports game.

[0065] FIG. 2 is a block diagram of a device (100) according to one embodiment of the present disclosure. In one embodiment, the device (100) may include one or more processors (210) and / or one or more memories (220) as elements. In one embodiment, at least one of the elements of the device (100) may be omitted, or another element may be added to the device (100). In one embodiment, additionally or alternatively, some of the elements may be implemented in an integrated manner or implemented as a single or multiple entities. One or more processors (210) may be referred to as a processor (210). The expression “processor (210)” may mean a set of one or more processors, unless the context clearly indicates otherwise. One or more memories (220) may be referred to as a memory (220). The expression “memory (220)” may mean a set of one or more memories, unless the context clearly indicates otherwise. At least some of the components inside / outside the device (100) are connected to each other through a bus, GPIO (general purpose input / output), SPI (serial peripheral interface), MIPI (mobile industry processor interface), etc., and can exchange information (data, signals, etc.).

[0066] In one embodiment, the processor (210) may control at least one component of the device (100) connected to the processor (210) by executing instructions (e.g., code, software, program, etc.). In addition, the processor (210) may perform various operations such as calculations, processing, data generation, and processing related to the present disclosure. In addition, the processor (210) may load data, etc. from the memory (220) or store data, etc. in the memory (220). For example, the processor (210) may obtain at least one of movement data (180) regarding the movement of an object in a target sports game or game situation data (170) indicating a game situation of the target sports game, and may detect an abnormality in the target sports game based on the movement data (180) or the game situation data (170), and generate analysis data (190) indicating the abnormality.

[0067] In one embodiment, the memory (220) can store various data. The data stored in the memory (220) is data acquired, processed, or used by at least one component of the device (100), and may include instructions (e.g., code, software, programs, etc.). The memory (220) may include volatile and / or non-volatile memory. The instructions or programs are software stored in the memory (220), and may include an operating system for controlling the resources of the device (100), applications, and / or middleware that provides various functions to applications so that the applications can utilize the resources of the device (100). For example, the memory (220) may store instructions that cause the processor (210) to perform operations when executed by the processor (210). The memory (220) may store image data (160), game situation data (170), movement data (180), analysis data (190), etc. of a target sports game. Additionally, the memory (220) may store image data, game situation data, movement data, analysis data, etc. of one or more other sports events. In one embodiment, the processor (210) may control a communication circuit to acquire information from another server or device. The information thus acquired may also be stored in the memory (220).

[0068] In one embodiment, the device (100) may further include a communication circuit (230). The communication circuit (230) may be omitted from the device (100) depending on the embodiment. The communication circuit (230) may perform wireless or wired communication between the device (100) and another server, or between the device (100) and another device. For example, the communication circuit (230) may perform wireless communication according to a method such as eMBB (enhanced Mobile Broadband), URLLC (Ultra Reliable Low-Latency Communications), MMTC (Massive Machine Type Communications), LTE (Long-Term Evolution), LTE-A (LTE Advance), NR (New Radio), UMTS (Universal Mobile Telecommunications System), GSM (Global System for Mobile communications), CDMA (Code Division Multiple Access), WCDMA (Wideband CDMA), WiBro (Wireless Broadband), WiFi (Wireless Fidelity), Bluetooth (Bluetooth), NFC (Near Field Communication), GPS (Global Positioning System), or GNSS (Global Navigation Satellite System). For example, the communication circuit (230) may perform wired communication according to a method such as USB (Universal Serial Bus), HDMI (High Definition Multimedia Interface), RS-232 (Recommended Standard-232), or POTS (Plain Old Telephone Service).For example, the communication circuit (230) can be used to perform communication with a camera group (130) or an external device (140, 150), etc. The communication circuit (230) can be implemented as a circuit or chip configured to perform data transmission and reception.

[0069] In one embodiment, the device (100) may further include an input / output interface. The input / output interface may be omitted from the device (100) depending on the embodiment. For example, the input / output interface may include an input device and / or an output device. The input device may receive various information from a user and transmit the input information to at least one component of the device (100). The output device may receive various information output by at least one component of the device (100) and provide (display) the information to the user in an audiovisual form. For example, the input device may include a mouse, a keyboard, a touch pad, etc. For example, the output device may include a display, a projector, a hologram, etc.

[0070] In one embodiment, the device (100) may be a device of various forms. For example, the device (100) may be a computer device, a back-end server, a front-end server, a portable communication device, a portable multimedia device, a wearable device, a device according to a combination of the aforementioned devices, or a chip, board, circuit, etc. within the aforementioned devices. However, the device (100) of the present disclosure is not limited to the aforementioned devices.

[0071] FIG. 3 is a diagram illustrating a process for generating location data (125) of an object (120) according to one embodiment of the present disclosure. In one embodiment, the processor (210) may receive image data (160) including an image set in which an image of a target sporting event is divided into frames from a camera group (130), and generate frame-by-frame location data (125) of an object (120) within a game space (110) of the target sporting event based on the image set. For convenience of explanation, it is assumed that the object (120) of FIG. 3 is a player of the target sporting event.

[0072] In one embodiment, the processor (210) may determine frame-by-frame three-dimensional coordinates of an object (120) from a plurality of images constituting an image set, and generate frame-by-frame position data (125) including the determined three-dimensional coordinates and the corresponding time. Here, the three-dimensional coordinates may be composed of X (x-axis), Y (y-axis), and Z (z-axis) values ​​in pixel units.

[0073] For example, the processor (210) may select one or more two-dimensional images corresponding to a specific frame (time) among a plurality of two-dimensional images in an image set, and determine two-dimensional coordinates of an object (120) in each of the selected one or more two-dimensional images.

[0074] For example, the two-dimensional coordinates of the object (120) may be one or more two-dimensional coordinates forming an outline of the object (120) within a two-dimensional image, or may be a center point of one or more such two-dimensional coordinates.

[0075] For example, the two-dimensional coordinates of the object (120) may be one of one or more two-dimensional coordinates indicating a specific part of the object (120) or a center point obtained by averaging one or more two-dimensional coordinates. As a specific example, if the target sport is a ball game, the specific part of the object (120) may be a part where the ball and the object (120) come into direct or indirect contact. That is, if the target sport is table tennis, the specific part of the object (120) may be a hand holding a racket. Alternatively, if the target sport is a non-ball game, the specific part of the object (120) may be a part that the object (120) mainly uses during the target sport. That is, if the target sport is boxing, the specific part of the object (120) may be a hand that the object (120) mainly uses during the target sport.

[0076] For example, the processor (210) can determine the three-dimensional coordinates of the object (120) corresponding to a specific frame (time) based on the two-dimensional coordinates of the object (120) in each of one or more images.

[0077] Meanwhile, the processor (210) can determine the three-dimensional coordinates of the object (120) from the image set using various computer vision algorithms, and the present disclosure is not limited thereto.

[0078] FIG. 4 is a diagram illustrating a process for generating movement data (180) of an object (120) according to one embodiment of the present disclosure. In one embodiment, the processor (210) may generate movement data (180) regarding the movement of an object (120) based on frame-by-frame position data (125) of the object (120) within the game space (110) of the target sports game.

[0079] In one embodiment, the movement data (180) of the object (120) may include indicators indicating the movement of the object (120), such as the direction of movement, the speed of movement, the time of movement (e.g., t seconds), and the movement trajectory of the object (120). Here, each indicator included in the movement data (180) may be numerical data or categorical data.

[0080] For example, movement data (180) of an object (120) may include at least one of a movement direction, a movement speed, a movement time, or a movement trajectory regarding the object (120) moving from a first location (X, Y, Z) to a second location (X', Y', Z').

[0081] FIG. 5 is a diagram illustrating a process of generating movement data (180) of an object (120) according to one embodiment of the present disclosure. In one embodiment, the processor (210) may generate movement data (180) regarding the movement of a specific portion of the object (120) based on frame-by-frame position data (125) of a specific portion of the object (120) within the game space (110) of the target sports game.

[0082] In one embodiment, movement data (180) of an object (120) may include indicators indicating movement of a specific part of the object (120), such as a movement direction, movement speed, movement time (e.g., t seconds), and movement trajectory of a specific part of the object (120). Here, each indicator included in the movement data (180) may be numerical data or categorical data.

[0083] For example, movement data (180) of an object (120) may include at least one of a movement direction, a movement speed, a movement time, or a movement trajectory regarding a specific part of the object (120) moving from a first position (X, Y, Z) to a second position (X', Y', Z').

[0084] FIG. 6 is a diagram illustrating a process for generating location data (125) of an object (120) according to one embodiment of the present disclosure. In one embodiment, the processor (210) may receive image data (160) including an image set in which an image of a target sporting event is divided into frames from a camera group (130), and generate frame-by-frame location data (125) of an object (120) within a game space (110) of the target sporting event based on the image set. For convenience of explanation, the object (120) of FIG. 6 is assumed to be a ball of the target sporting event.

[0085] In one embodiment, the processor (210) may determine frame-by-frame three-dimensional coordinates of an object (120) from a plurality of images constituting an image set, and generate frame-by-frame position data (125) including the determined three-dimensional coordinates and the corresponding time. Here, the three-dimensional coordinates may be composed of X (x-axis), Y (y-axis), and Z (z-axis) values ​​in pixel units.

[0086] For example, the processor (210) may select one or more two-dimensional images corresponding to a specific frame (time) among a plurality of two-dimensional images in an image set, and determine two-dimensional coordinates of an object (120) in each of the selected one or more two-dimensional images.

[0087] For example, the two-dimensional coordinates of the object (120) may be any one of one or more two-dimensional coordinates that constitute the outline of the object (120) within the two-dimensional image, or may be the midpoint of one or more such two-dimensional coordinates.

[0088] For example, the processor (210) can determine the three-dimensional coordinates of the object (120) corresponding to a specific frame (time) based on the two-dimensional coordinates of the object (120) in each of one or more images.

[0089] Meanwhile, the processor (210) can determine the three-dimensional coordinates of the object (120) from the image set using various computer vision algorithms. However, the present disclosure is not limited to using computer vision algorithms for such determination.

[0090] FIG. 7 is a diagram illustrating a process of generating movement data (180) of an object (120) according to one embodiment of the present disclosure. In one embodiment, the processor (210) may generate movement data (180) regarding the movement of an object (120) based on frame-by-frame position data (125) of the object (120) within the game space (110) of the target sports game.

[0091] In one embodiment, the movement data (180) of the object (120) may include indicators indicating the movement of the object (120), such as the direction of movement, speed of movement, movement time (e.g., t seconds), movement trajectory, spin direction, and number of spins of the object (120). Here, each indicator included in the movement data (180) may be numerical data or categorical data.

[0092] For example, movement data (180) of an object (120) may include at least one of a movement direction, a movement speed, a movement time, a movement trajectory, a spin direction, or a spin count regarding the movement of the object (120) from a first position (X, Y, Z) to a second position (X', Y', Z').

[0093] Meanwhile, the processor (210) may obtain movement data (180) of an object (120) of a target sport depending on the type of the target sport. For example, if the target sport is a ball game, the processor (210) may generate frame-by-frame position data (125) of a player (120) and frame-by-frame position data (125) of a ball (120) of the target sport, and may generate movement data (180) of the player and movement data (180) of the ball based on the frame-by-frame position data (125) of the player (120) and frame-by-frame position data (125) of the ball (120), respectively. Alternatively, if the target sport is a non-ball game, the processor (210) may generate frame-by-frame position data (125) of a player of the target sport, and generate movement data (180) of the player based on the frame-by-frame position data (125) of the player.

[0094] FIG. 8 is a diagram illustrating a process of obtaining analysis data (190) using a learning model (800) trained to detect the presence or absence of an abnormality in a sports game according to one embodiment of the present disclosure.

[0095] In one embodiment, the learning model (800) may be a model that is trained to detect the presence or absence of an anomaly in a sporting event using learning data (810) for other sporting events during the learning process, and generates output data (830) indicating the presence or absence of an anomaly in the target sporting event from input data (820) for the target sporting event during the inference process.

[0096] For example, the processor (210) may input at least one of movement data (180) of an object (120) in a target sports game or game situation data (170) of the target sports game as input data (820) to the learning model (800), and obtain analysis data (190) indicating the presence or absence of an abnormality in the target sports game as output data (830) of the learning model (800).

[0097] In one embodiment, the learning model (800) may be a model trained using movement patterns (hereinafter, "first movement patterns") of an object (120) in one or more other sporting events as training data (810). Here, the one or more other sporting events may be sporting events distinct from the target sporting event. The learning model (800) that has trained the first movement patterns of the object (120) in one or more other sporting events may detect whether the object (120) exhibits abnormal behavior in the target sporting event.

[0098] For example, the first movement pattern may refer to statistical data on indicators indicating the movement of the object (120), such as the direction of movement, speed of movement, time of movement, and movement trajectory of the object (120) in one or more other sports games. As a specific example, the first movement pattern may include the average movement speed, average movement trajectory, and the like of the object (120) in one or more other sports games.

[0099] For example, the first movement pattern may mean statistical data on indicators indicating the movement of the object (120), such as the direction of movement, movement speed, movement time, and movement trajectory of the object (120), at a specific point in time in one or more different sports games.

[0100] For example, the first movement pattern may mean statistical data on indicators indicating the movement of the object (120), such as the direction of movement, movement speed, movement time, and movement trajectory of the object (120) in a specific game situation of one or more different sports games.

[0101] Meanwhile, in the inference process, the processor (210) inputs movement data (180) of an object (120) in a target sports game into a learned learning model (800) as described above, so that analysis data (190) indicating the presence or absence of an abnormality in the target sports game can be obtained as an output of the learning model (800).

[0102] In one embodiment, the processor (210) may perform preprocessing on input data (820) of the learning model (800). For example, the processor (210) may extract a movement pattern (hereinafter, “second movement pattern”) of an object (120) in a target sports event based on movement data (180) of the object (120) in the target sports event. The processor (210) may utilize the second movement pattern as input data (820) for the learning model (800).

[0103] For example, the second movement pattern may mean statistical data on indicators indicating the actual movement of the object (120), such as the direction of movement, movement speed, movement time, and movement trajectory, in the target sports game.

[0104] For example, the second movement pattern may mean statistical data on indicators indicating the movement of the object (120), such as the direction of movement, movement speed, movement time, and movement trajectory of the object (120) at a specific point in the target sports game.

[0105] For example, the second movement pattern may mean statistical data on indicators indicating the movement of the object (120), such as the direction of movement, movement speed, movement time, and movement trajectory of the object (120) in a specific game situation of the target sports game.

[0106] For example, the processor (210) may obtain analysis data (190) from a second movement pattern using a learning model (800) that has learned a first movement pattern of an object (120) in one or more other sports games. Specifically, the processor (210) may input the second movement pattern as input data (820) into the learning model (800) and obtain analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) may generate expected movement data of the object (120) in the target sports game based on the first movement pattern learned in advance. Here, the expected movement data may include indicators indicating the expected movement of the object (120), such as the expected movement direction, expected movement speed, expected movement time, and expected movement trajectory of the object (120) in the target sports game. The learning model (800) can generate analysis data (190) indicating the presence or absence of an abnormality in a target sports game based on the second movement pattern and expected movement data.

[0107] For example, the analysis data (190) may include at least one of a first movement pattern, a second movement pattern, a comparison result between the first movement pattern and the second movement pattern, expected movement data, or a comparison result between the second movement pattern and the expected movement data. In addition, the analysis data (190) may also include an indicator or notification indicating that an abnormality has been detected in the target sporting event. Here, whether the analysis data (190) includes an indicator or notification indicating that an abnormality has been detected in the target sporting event may be determined based on the comparison result between the first movement pattern and the second movement pattern or the comparison result between the second movement pattern and the expected movement data.

[0108] In one embodiment, the learning model (800) may be a model trained using a match situation pattern (hereinafter, "first match situation pattern") of a different sporting event as training data (810). Here, the "other sporting event" may refer to a sport that is identical to the target sporting event, includes at least some of the same players (teams) participating in the sporting event, and is held at a different time from the target sporting event. The learning model (800) that has trained the first match situation pattern of a different sporting event can be used to detect abnormal match situations, such as a persistent negative match flow for a specific team, in the target sporting event by comparing it with other sporting events.

[0109] For example, the first game situation pattern may refer to statistical data on indicators indicating the game situation of a given sporting event, such as the game time, each team's score, each team's game control time, each team's rule violations (e.g., fouls, warnings, red cards), and referee decisions. Here, game control time may refer to the time a specific team is in a superior position over the opposing team. For example, game control time may refer to the duration of a specific team's score being higher than the opposing team's, or the time a specific team's player is in possession of the ball in the opposing team's territory. Furthermore, referee decisions may refer to the referee's decisions regarding rule violations, goals, etc. As a specific example, the first game situation pattern may include the average game control time of a specific team in a different sporting event. As a specific example, the first game situation pattern may include the decision tendencies of a specific referee in a different sporting event.

[0110] For example, in the inference process, the processor (210) inputs the game situation data (170) of the target sports game into the learned learning model (800) as described above, and can obtain analysis data (190) indicating the presence or absence of an abnormality in the target sports game as the output of the learning model (800).

[0111] For example, the processor (210) may preprocess input data (820) of the learning model (800). For example, the processor (210) may extract a match situation pattern (hereinafter, “second match situation pattern”) of the target sports game based on match situation data (170) of the target sports game. The processor (210) may utilize the second match situation pattern as input data (820) for the learning model (800).

[0112] For example, the second game situation pattern may refer to statistical data on indicators indicating the actual game situation of the target sport game, such as the game time of the target sport game, the score of each team, the game time of each team, the rule violations (e.g., fouls, warnings, red cards), and the results of referee decisions.

[0113] For example, the processor (210) can obtain analysis data (190) from a second game situation pattern using a learning model (800) that has learned a first game situation pattern of another sports game. Specifically, the processor (210) can input the second game situation pattern as input data (820) into the learning model (800) and obtain analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) can generate expected game situation data of the target sports game based on the first game situation pattern learned in advance. Here, the expected game situation data may mean statistical data on indicators indicating the expected game situation of the target sports game, such as the expected score of each team in the target sports game, the expected game leading time of each team, the expected rule violations (e.g., fouls, warnings, red cards) of each team, and the expected referee decision results.

[0114] For example, the analysis data (190) may include at least one of a first game situation pattern, a second game situation pattern, a comparison result between the first game situation pattern and the second game situation pattern, expected game situation data, or a comparison result between the second game situation pattern and the expected game situation data. In addition, the analysis data (190) may also include an indicator or notification indicating that an abnormality has been detected in the target sport game. Here, whether the analysis data (190) includes an indicator or notification indicating that an abnormality has been detected in the target sport game may be determined based on the comparison result between the first game situation pattern and the second game situation pattern or the comparison result between the second game situation pattern and the expected game situation data.

[0115] In one embodiment, the learning model (800) may be a model trained using the first game situation patterns of each of a plurality of different sports events as training data (810). Here, the plurality of different sports events may refer to games that are identical to the target sport event, have the same event, specific players (teams), or referees participating in the game, and are played at different times from the target sport event. The learning model (800) that has learned the first game situation patterns of each of the plurality of different sports events can be used to detect whether there are unique game situations unique to the target sport event compared to the plurality of other sports events.

[0116] For example, the first game situation pattern of each of multiple different sports games may mean statistical data on indicators indicating the game situation of the corresponding sports game, such as the game time, each team's score, each team's game time, each team's rule violations (e.g., fouls, warnings, red cards), and referee decisions.

[0117] For example, the processor (210) can input game situation data (170) of a target sports game into a learned learning model (800) as described above, and obtain analysis data (190) indicating the presence or absence of an abnormality in the target sports game as an output of the learning model (800).

[0118] Meanwhile, the processor (210) may also preprocess the input data (820) of the learning model (800). For example, the processor (210) may extract a second game situation pattern of the target sports game based on game situation data (170) of the target sports game.

[0119] For example, the second game situation pattern may refer to statistical data on indicators indicating the actual game situation of the target sport game, such as the game time of the target sport game, the score of each team, the game time of each team, the rule violations (e.g., fouls, warnings, red cards), and the results of referee decisions.

[0120] For example, the processor (210) can obtain analysis data (190) from a second game situation pattern using a learning model (800) that has learned the first game situation pattern of each of a plurality of different sports games. Specifically, the processor (210) can input the second game situation pattern as input data (820) into the learning model (800) and obtain analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) can generate expected game situation data of a target sports game based on the first game situation patterns of each of a plurality of different sports games that have been learned in advance. Here, the expected game situation data may mean statistical data on indicators indicating the expected game situation of the target sports game, such as the expected score of each team in the target sports game, the expected game leading time of each team, the expected rule violations (e.g., fouls, warnings, red cards) of each team, and the expected referee decision results.

[0121] For example, the analysis data (190) may include at least one of a first game situation pattern, a second game situation pattern of each of a plurality of different sporting events, a comparison result between the first game situation pattern and the second game situation pattern of each of a plurality of different sporting events, expected game situation data, or a comparison result between the second game situation pattern and the expected game situation data.

[0122] FIG. 9 is a diagram illustrating a process for acquiring analysis data (190) according to one embodiment of the present disclosure. In one embodiment, the processor (210) may acquire analysis data (190) indicating the presence or absence of an abnormality in the target sports game based on at least one of movement data (180) of an object (120) within a game space (110) of the target sports game or game situation data (170) of the target sports game. For convenience of explanation, it is assumed that the object (120) in FIG. 9 is a player of the target sports game.

[0123] In one embodiment, the analysis data (190) may include at least one of a first movement pattern (not shown), a second movement pattern (910), a comparison result between the first movement pattern (not shown) and the second movement pattern (910), expected movement data (920), or a comparison result between the second movement pattern (910) and the expected movement data (920). For convenience of explanation, it is assumed in FIG. 9 that the analysis data (190) includes the second movement pattern (910) and the expected movement data (920).

[0124] In one embodiment, the analysis data (190) may include an indicator or notification indicating that an abnormality has been detected in the target sporting event. To this end, the processor (210) may determine whether to include an indicator or notification indicating that an abnormality has been detected in the target sporting event in the analysis data (190) based on the comparison results between the second movement pattern (910) and the expected movement data (920).

[0125] For example, if the difference between the actual movement trajectory of the object (120) according to the second movement pattern (910) and the expected movement trajectory of the object (120) according to the expected movement data (920) is outside a predetermined range, the processor (210) may decide to include an indicator or notification indicating that an abnormality has been detected in the target sports game in the analysis data (190).

[0126] FIG. 10 is a diagram illustrating a process for obtaining analysis data (190) according to one embodiment of the present disclosure. In one embodiment, the processor (210) may obtain analysis data (190) indicating the presence or absence of an abnormality in the target sports game based on at least one of movement data (180) of an object (120) within a game space (110) of the target sports game or game situation data (170) of the target sports game. For convenience of explanation, the object (120) in FIG. 10 is assumed to be a ball in the target sports game.

[0127] In one embodiment, the analysis data (190) may include at least one of a first movement pattern (not shown), a second movement pattern (1010), a comparison result between the first movement pattern (not shown) and the second movement pattern (1010), expected movement data (1020), or a comparison result between the second movement pattern (1010) and the expected movement data (1020). For convenience of explanation, FIG. 10 assumes that the analysis data (190) includes the second movement pattern (1010) and the expected movement data (1020).

[0128] In one embodiment, the analysis data (190) may include an indicator or notification indicating that an abnormality has been detected in the target sporting event. To this end, the processor (210) may determine whether to include an indicator or notification indicating that an abnormality has been detected in the target sporting event in the analysis data (190) based on the comparison results between the second movement pattern (1010) and the expected movement data (1020).

[0129] For example, if the difference between the actual movement speed of the object (120) according to the second movement pattern (1010) and the expected movement speed of the object (120) according to the expected movement data (1020) is outside a predetermined range, the processor (210) may decide to include an indicator or notification in the analysis data (190) indicating that an abnormality has been detected in the target sporting event.

[0130] FIG. 11 is a diagram illustrating a process for obtaining analysis data (190) according to one embodiment of the present disclosure. In one embodiment, the processor (210) may obtain analysis data (190) indicating the presence or absence of an abnormality in the target sports game based on game situation data (170) of the target sports game.

[0131] In one embodiment, the analysis data (190) may include at least one of a first match situation pattern (1110), a second match situation pattern (1120), a comparison result between the first match situation pattern (1110) and the second match situation pattern (1120), expected match situation data (1130), or a comparison result between the second match situation pattern (1120) and the expected match situation data (1130). For convenience of explanation, FIG. 11 assumes that the analysis data (190) includes the first match situation pattern (1110), the second match situation pattern (1120), and the expected match situation data (1130).

[0132] In one embodiment, the analysis data (190) may include an indicator or notification indicating that an abnormality has been detected in the target sporting event. To this end, the processor (210) may determine whether to include an indicator or notification indicating that an abnormality has been detected in the target sporting event in the analysis data (190) based on at least one of the comparison results of the first game situation pattern (1110) and the second game situation pattern (1120) or the comparison results of the second game situation pattern (1120) and the expected game situation data (1130).

[0133] For example, if the difference between a game situation indicator of a specific team at a specific point in time (e.g., t4) according to a first game situation pattern (1110) and a game situation indicator of the corresponding team at a specific point in time (e.g., t4) according to a second game situation pattern (1120) is outside a predetermined range, the processor (210) may decide to include in the analysis data (190) an indicator or notification indicating that an abnormality has been detected at a specific point in time (e.g., t4) of the target sport game or in a sport game that was played at a specific point in time (e.g., t4).

[0134] For example, if the difference between a game situation indicator of a specific team at a specific point in time (e.g., t4) according to the second game situation pattern (1120) and a game situation indicator of the corresponding team at a specific point in time (e.g., t4) according to the expected game situation data (1130) is outside a predetermined range, the processor (210) may decide to include in the analysis data (190) an indicator or notification indicating that an abnormality has been detected at a specific point in time (e.g., t4) of the target sporting event or in a sporting event that was played at a specific point in time (e.g., t4).

[0135] FIG. 12 is a diagram illustrating a method (1200) for detecting the presence or absence of an abnormality in a sports game according to one embodiment of the present disclosure. The method (1200) of the present disclosure can be performed by a device (100).

[0136] In step S1210, the processor (210) can obtain at least one of movement data (180) regarding the movement of an object (120) in a target sports game or game situation data (170) indicating a game situation of the target sports game.

[0137] In one embodiment, in obtaining movement data (180), the processor (210) may receive image data (160) for a target sporting event from a camera group (130) including one or more cameras (131, 132, 133, 134, 135, 136), and generate movement data (180) based on the image data (160).

[0138] In one embodiment, in generating movement data (180), the processor (210) may generate frame-by-frame position data (125) of an object (120) of a target sporting event based on an image set including one or more images divided into frame units within image data (160), and generate movement data (180) of the object (120) based on the frame-by-frame position data (125). This may refer to the descriptions of FIGS. 3 to 7.

[0139] In one embodiment, in obtaining game situation data (170), the processor (210) may receive game situation data (170) from an external device (140).

[0140] In one embodiment, in obtaining game situation data (170), the processor (210) may receive image data (160) from the camera group (130) and generate game situation data (170) based on the image data (160).

[0141] At step S1220, the processor (210) can detect whether there is an abnormality in the target sports game based on at least one of movement data (180) or game situation data (170) and obtain analysis data (190) indicating whether there is an abnormality.

[0142] In one embodiment, when acquiring analysis data (190), the processor (210) may input at least one of movement data (180) or game situation data (170) into a learning model (800) trained to detect the presence or absence of an abnormality in a sports game, and acquire analysis data (190) as an output of the learning model (800). This may be referred to the description of FIG. 8.

[0143] Additionally, the processor (210) can transmit at least one of movement data (180), game situation data (170), or analysis data (190) to an external device (150).

[0144] The method (1200) proposed in this disclosure is not limited to sports, and can be applied to various sports, including ball games and non-ball games. Examples used to explain the method (1200), such as movement data (180), match situation data (170), and analysis data (190), can be modified to suit the characteristics of each sport.

[0145] Below, an example of a soccer game is described.

[0146] The processor (210) can perform the method (1200) to detect the presence or absence of an abnormality in a soccer game.

[0147] In step S1210, the processor (210) may obtain at least one of movement data (180) regarding the movement of an object (120) in a target soccer game or game situation data (170) indicating a game situation in the target soccer game. In the target soccer game, the object (120) may be a player and a ball.

[0148] For example, movement data (180) may include indicators indicating the movement of the player in the target soccer game, such as the player's movement direction, movement speed, movement time, movement trajectory, reaction time, posture, dribble, pass, and shooting.

[0149] For example, movement data (180) may include indicators indicating the movement of the ball in the target soccer game, such as the direction of movement, movement speed, movement time, movement trajectory, vertical launch angle, left and right direction angle, and number of spins.

[0150] For example, the match situation data (170) may include indicators indicating the match situation of the target soccer match, such as events that occurred in the target soccer match (e.g., corner kick, free kick, penalty kick), rule violations by each team (e.g., foul, offside, caution, red card), each team's score, each team's game leading time, and referee decision results.

[0151] For example, in acquiring movement data (180), the processor (210) may receive image data (160) of a target soccer game from a camera group (130). In this case, the processor (210) may generate frame-by-frame position data (125) of a player and frame-by-frame position data (125) of a ball in the target soccer game based on an image set included in the image data (160). The processor (210) may generate movement data (180) of a player based on the frame-by-frame position data (125) of the player. The processor (210) may generate movement data (180) of a ball based on the frame-by-frame position data (125) of the ball.

[0152] For example, in obtaining game situation data (170), the processor (210) may receive game situation data (170) of a target soccer game from an external device (140). Additionally or alternatively, the processor (210) may generate game situation data (170) based on image data (160) received from a camera group (130).

[0153] At step S1220, the processor (210) can obtain analysis data (190) indicating the presence or absence of an abnormality in the target soccer game based on at least one of movement data (180) or game situation data (170).

[0154] For example, the processor (210) may input at least one of movement data (180) or game situation data (170) into a learning model (800) trained to detect the presence or absence of an abnormality in a soccer game, and obtain analysis data (190) indicating the presence or absence of an abnormality in the target soccer game as an output of the learning model (800).

[0155] For example, the learning model (800) may be a model learned using a first movement pattern of a player in one or more other soccer games as learning data (810). Here, the first movement pattern may mean statistical data on indicators indicating the movement of the player, such as the movement direction, movement speed, movement time, movement trajectory, reaction time, posture, dribbling, passing, and shooting of the player in one or more other soccer games. The processor (210) may extract a second movement pattern of the player from the movement data (180). Here, the second movement pattern may be statistical data on indicators indicating the actual movement of the player, such as the movement direction, movement speed, movement time, movement trajectory, reaction time, posture, dribbling, passing, and shooting of the player in the target soccer game. The processor (210) may obtain analysis data (190) from the second movement pattern using the learning model (800) that learned the first movement pattern of the player in one or more other soccer games. The processor (210) can input the second movement pattern as input data (820) into the learning model (800) and obtain analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) can generate expected movement data of a player in a target soccer game based on the first movement pattern learned in advance. Here, the expected movement data can include indicators indicating expected movements of the corresponding player, such as expected movement direction, expected movement speed, expected movement time, expected movement trajectory, expected reaction time, expected posture, and expected dribble. The learning model (800) can generate analysis data (190) indicating the presence or absence of an abnormality in the target soccer game based on the second movement pattern and the expected movement data.For example, the analysis data (190) may include at least one of a first movement pattern, a second movement pattern, a comparison result between the first movement pattern and the second movement pattern, expected movement data, or a comparison result between the second movement pattern and the expected movement data. In addition, the analysis data (190) may also include an indicator or notification indicating that an abnormal behavior of a specific player has been detected in the target soccer game. Here, based on the comparison result between the first movement pattern and the second movement pattern or the comparison result between the second movement pattern and the expected movement data, it may be determined whether the analysis data (190) includes an indicator or notification indicating that an abnormal behavior of a specific player has been detected in the target soccer game. If, as a result of comparing the first movement pattern and the second movement pattern, the change in the player's reaction time or movement trajectory is outside a predetermined range, the processor (210) may determine to include in the analysis data (190) an indicator or notification indicating that the player is performing abnormal behavior in the target soccer game. If, as a result of comparing the first movement pattern and the second movement pattern, the number of dribble attempts or the dribble success rate of the player is reduced by a predetermined value or more, the processor (210) may determine to include in the analysis data (190) an indicator or notification indicating that the player is performing abnormal behavior in the target soccer game. If, as a result of comparing the second movement pattern and the expected movement data, the difference between the player's expected movement direction and the actual movement direction is outside a predetermined range, the processor (210) may determine to include in the analysis data (190) an indicator or notification indicating that the player is performing abnormal behavior in the target soccer game.If, as a result of comparing the expected movement data and the second movement pattern, a player was expected to shoot at a certain point in time but did not actually shoot, the processor (210) may decide to include an indicator or notification in the analysis data (190) indicating that there is abnormal behavior by the player in the target soccer game.

[0156] For example, the learning model (800) may be a model trained using a first game situation pattern of another soccer game as learning data (810). For example, the first game situation pattern may refer to statistical data on indicators indicating the game situation of the soccer game, such as events that occurred in another soccer game, rule violations by each team, each team's score, each team's game-dominant time, and referee decision results. The processor (210) may input the game situation data (170) of the target soccer game into the learning model (800) to obtain analysis data (190) indicating the presence or absence of an abnormality in the target soccer game as an output of the learning model (800). The processor (210) may extract a second game situation pattern of the target soccer game based on the game situation data (170) of the target soccer game. The second game situation pattern may refer to statistical data on indicators indicating the actual game situation of the target soccer game, such as events that occurred in the target soccer game, rule violations by each team, each team's score, each team's game-dominant time, and referee decision results. The processor (210) can obtain analysis data (190) from a second game situation pattern using a learning model (800) that has learned a first game situation pattern of another soccer game.

[0157] Specifically, the processor (210) can input the second game situation pattern as input data (820) into the learning model (800) and obtain the analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) can generate expected game situation data of the target soccer game based on the first game situation pattern learned in advance. Here, the expected game situation data can include indicators indicating the expected game situation of the target soccer game, such as expected events, expected rule violations of each team, expected scores of each team, expected game leading time of each team, and expected decision results of the referee. The learning model (800) can generate analysis data (190) indicating the presence or absence of an abnormality in the target soccer game based on the second game situation pattern and the expected game situation data. For example, the analysis data (190) may include at least one of a first game situation pattern, a second game situation pattern, a comparison result between the first game situation pattern and the second game situation pattern, expected game situation data, or a comparison result between the second game situation pattern and the expected game situation data. In addition, the analysis data (190) may include an indicator or notification indicating that an abnormal game situation exists in the target soccer game. Here, based on the comparison result between the first game situation pattern and the second game situation pattern or the comparison result between the second game situation pattern and the expected game situation data, it may be determined whether the analysis data (190) includes an indicator or notification indicating that an abnormal game situation exists in the target soccer game. If, as a result of comparing the second game situation pattern and the expected game situation data, the referee's expected decision and the actual decision for a specific situation are different, the processor (210) may determine to include an indicator or notification indicating that an abnormal game situation exists in the analysis data (190).

[0158] For example, the learning model (800) may be a model learned using the first game situation pattern of each of a plurality of different soccer games as learning data (810). The processor (210) may input the game situation data (170) of the target soccer game into the learning model (800) to obtain analysis data (190) indicating the presence or absence of an abnormality in the target soccer game as an output of the learning model (800). The processor (210) may extract the second game situation pattern of the target soccer game based on the game situation data (170) of the target soccer game. The processor (210) may obtain analysis data (190) from the second game situation pattern using the learning model (800) that has learned the first game situation pattern of each of a plurality of different soccer games. Specifically, the processor (210) may input the second game situation pattern as input data (820) into the learning model (800) and obtain the analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) may generate expected game situation data of the target soccer game based on the first game situation pattern of each of a plurality of other soccer games learned in advance. Here, the expected game situation data may include an indicator indicating the expected game situation of the target soccer game, such as a win / loss prediction or a score prediction. The learning model (800) may generate analysis data (190) indicating the presence or absence of an abnormality in the target soccer game based on the second game situation pattern and the expected game situation data. For example, the analysis data (190) may include at least one of the first game situation pattern of each of the plurality of other soccer games, the second game situation pattern, a comparison result between the first game situation pattern and the second game situation pattern of each of the plurality of other soccer games, the expected game situation data, or a comparison result between the second game situation pattern and the expected game situation data.Additionally, the analysis data (190) may include an indicator or notification indicating that an abnormal game situation exists in the target soccer game. Here, whether the analysis data (190) includes an indicator or notification indicating that an abnormal game situation exists in the target soccer game may be determined based on a comparison result between the first game situation pattern and the second game situation pattern of each of a plurality of other soccer games, or a comparison result between the second game situation pattern and the expected game situation data. If the comparison result of the second game situation pattern and the expected game situation data indicates that a specific team lost by a larger score than expected, the processor (210) may determine to include an indicator or notification indicating that an abnormal game situation exists by the specific team in the target soccer game in the analysis data (190).

[0159] Below, an example of a basketball game is described.

[0160] The processor (210) can perform the method (1200) to detect the presence or absence of an abnormality in a basketball game.

[0161] In step S1210, the processor (210) may obtain at least one of movement data (180) regarding the movement of an object (120) in a target basketball game or game situation data (170) indicating a game situation in the target basketball game. In the target basketball game, the object (120) may be a player and a ball.

[0162] For example, movement data (180) may include indicators indicating the movement of the player in the target basketball game, such as the player's movement direction, movement speed, movement time, movement trajectory, reaction time, posture, dribbling, passing, shooting, and rebounding.

[0163] For example, movement data (180) may include indicators indicating the movement of the ball in the target basketball game, such as the direction of movement, movement speed, movement time, movement trajectory, vertical launch angle, left and right direction angle, and number of spins.

[0164] For example, the game situation data (170) may include indicators indicating the game situation of the target basketball game, such as events that occurred in the target basketball game (e.g., free throws), rule violations by each team (e.g., fouls, ejections), scores by each team, game leading time by each team, and referee decision results.

[0165] For example, in acquiring movement data (180), the processor (210) may receive image data (160) of a target basketball game from the camera group (130). In this case, the processor (210) may generate frame-by-frame position data (125) of a player and frame-by-frame position data (125) of a ball in the target basketball game based on an image set included in the image data (160). The processor (210) may generate movement data (180) of a player based on the frame-by-frame position data (125) of the player. The processor (210) may generate movement data (180) of a ball based on the frame-by-frame position data (125) of the ball.

[0166] For example, in obtaining game situation data (170), the processor (210) may receive game situation data (170) of a target basketball game from an external device (140). Additionally or alternatively, the processor (210) may generate game situation data (170) based on image data (160) received from the camera group (130).

[0167] At step S1220, the processor (210) can obtain analysis data (190) indicating the presence or absence of an abnormality in the target basketball game based on at least one of movement data (180) or game situation data (170).

[0168] For example, the processor (210) may input at least one of movement data (180) or game situation data (170) into a learning model (800) trained to detect the presence or absence of an abnormality in a basketball game, and obtain analysis data (190) indicating the presence or absence of an abnormality in the target basketball game as an output of the learning model (800).

[0169] For example, the learning model (800) may be a model learned using a first movement pattern of a player in one or more other basketball games as learning data (810). Here, the first movement pattern may refer to statistical data on indicators indicating the movement of the player, such as the movement direction, movement speed, movement time, movement trajectory, reaction time, posture, dribble, pass, shooting, and rebound of the player in one or more other basketball games. The processor (210) may extract a second movement pattern of the player from the movement data (180). Here, the second movement pattern may be statistical data on indicators indicating the actual movement of the player, such as the movement direction, movement speed, movement time, movement trajectory, reaction time, posture, dribble, pass, shooting, and rebound of the player in the target basketball game. The processor (210) can obtain analysis data (190) from the second movement pattern using a learning model (800) that has learned the first movement pattern of a player in one or more other basketball games. The processor (210) can input the second movement pattern as input data (820) into the learning model (800) and obtain analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) can generate expected movement data of the player in the target basketball game based on the first movement pattern learned in advance. Here, the expected movement data can include indicators indicating the expected movement of the corresponding player, such as a movement direction, a movement speed, a movement time, a movement trajectory, a reaction time, a posture, a dribble, a pass, a shot, a rebound, etc. The learning model (800) can generate analysis data (190) that indicates the presence or absence of an abnormality in the target basketball game based on the second movement pattern and the expected movement data.For example, the analysis data (190) may include at least one of a first movement pattern, a second movement pattern, a comparison result between the first movement pattern and the second movement pattern, expected movement data, or a comparison result between the second movement pattern and the expected movement data. In addition, the analysis data (190) may also include an indicator or notification indicating that abnormal behavior of a specific player has been detected in the target basketball game. Here, based on the comparison result between the first movement pattern and the second movement pattern or the comparison result between the second movement pattern and the expected movement data, it may be determined whether the analysis data (190) includes an indicator or notification indicating that abnormal behavior of a specific player has been detected in the target basketball game. If the comparison result between the first movement pattern and the second movement pattern indicates that the shooting accuracy of the player has decreased by a predetermined value or more, the processor (210) may determine to include an indicator or notification indicating that abnormal behavior of the corresponding player has been detected in the analysis data (190). As a result of comparing the second movement pattern and the expected movement data, if the difference between the player's expected pass path and the actual pass path in a specific situation is greater than a predetermined range, the processor (210) may determine to include an indicator or notification in the analysis data (190) indicating that there is an abnormal behavior of the player in the target basketball game.

[0170] For example, the learning model (800) may be a model trained using a first game situation pattern of another basketball game as learning data (810). For example, the first game situation pattern may refer to statistical data on indicators indicating the game situation of the corresponding basketball game, such as events that occurred in another basketball game, rule violations by each team, each team's score, each team's game-dominant time, and referee decision results. The processor (210) may input game situation data (170) of the target basketball game into the learning model (800) to obtain analysis data (190) indicating the presence or absence of an abnormality in the target basketball game as an output of the learning model (800). The processor (210) may extract a second game situation pattern of the target basketball game based on the game situation data (170) of the target basketball game. The second game situation pattern may refer to statistical data on indicators indicating the actual game situation of the target basketball game, such as events that occurred in the target basketball game, rule violations by each team, each team's score, each team's game-dominant time, and referee decision results. The processor (210) can obtain analysis data (190) from a second game situation pattern using a learning model (800) that has learned a first game situation pattern of another basketball game. Specifically, the processor (210) can input the second game situation pattern as input data (820) into the learning model (800) and obtain analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) can generate expected game situation data of the target basketball game based on the first game situation pattern learned in advance. Here, the expected game situation data can include indicators indicating the expected game situation of the target basketball game, such as expected events, expected rule violations of each team, expected scores of each team, expected game leading time of each team, and expected decision results of the referee.The learning model (800) can generate analysis data (190) indicating whether there is an abnormality in the target basketball game based on the second game situation pattern and the expected game situation data. For example, the analysis data (190) can include at least one of the first game situation pattern, the second game situation pattern, the comparison result between the first game situation pattern and the second game situation pattern, the expected game situation data, or the comparison result between the second game situation pattern and the expected game situation data. In addition, the analysis data (190) can include an indicator or notification indicating that there is an abnormal game situation in the target basketball game. Here, whether the analysis data (190) includes an indicator or notification indicating that there is an abnormal game situation in the target basketball game can be determined based on the comparison result between the first game situation pattern and the second game situation pattern or the comparison result between the second game situation pattern and the expected game situation data. If, as a result of comparing the second game situation pattern and the expected game situation data, the referee's expected decision for a specific situation differs from the actual decision, the processor (210) may decide to include an indicator or notification in the analysis data (190) indicating that there is an abnormal game situation in the target basketball game.

[0171] For example, the learning model (800) may be a model learned using the first game situation pattern of each of a plurality of different basketball games as learning data (810). The processor (210) may input the game situation data (170) of the target basketball game into the learning model (800) to obtain analysis data (190) indicating the presence or absence of an abnormality in the target basketball game as an output of the learning model (800). The processor (210) may extract the second game situation pattern of the target basketball game based on the game situation data (170) of the target basketball game. The processor (210) may obtain analysis data (190) from the second game situation pattern using the learning model (800) that has learned the first game situation pattern of each of a plurality of different basketball games. Specifically, the processor (210) may input the second game situation pattern as input data (820) into the learning model (800) and obtain the analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) may generate predicted game situation data of the target basketball game based on the first game situation pattern of each of a plurality of other basketball games learned in advance. Here, the predicted game situation data may include an indicator indicating the predicted game situation of the target basketball game, such as a win / loss prediction or a score prediction. The learning model (800) may generate analysis data (190) indicating the presence or absence of an abnormality in the target basketball game based on the second game situation pattern and the predicted game situation data. For example, the analysis data (190) may include at least one of the first game situation pattern of each of the plurality of other basketball games, the second game situation pattern, a comparison result between the first game situation pattern and the second game situation pattern of each of the plurality of other basketball games, the predicted game situation data, or a comparison result between the second game situation pattern and the predicted game situation data.Additionally, the analysis data (190) may include an indicator or notification indicating that an abnormal game situation exists in the target basketball game. Here, whether the analysis data (190) includes an indicator or notification indicating that an abnormal game situation exists in the target basketball game may be determined based on a comparison result between the first game situation pattern and the second game situation pattern of each of a plurality of different basketball games, or a comparison result between the second game situation pattern and the expected game situation data. If the difference between the expected score and the actual score of a specific team is greater than or equal to a predetermined value as a result of comparing the second game situation pattern and the expected game situation data, the processor (210) may determine to include an indicator or notification indicating that an abnormal game situation exists by a specific team in the target basketball game in the analysis data (190).

[0172] Below, an example of a volleyball game is described.

[0173] The processor (210) can perform the method (1200) to detect the presence or absence of an abnormality in a volleyball game.

[0174] In step S1210, the processor (210) may obtain at least one of movement data (180) regarding the movement of an object (120) in a target volleyball match or match situation data (170) indicating a match situation in the target volleyball match. In the target volleyball match, the object (120) may be a player and a ball.

[0175] For example, movement data (180) may include indicators indicating the movement of the player in the target volleyball game, such as the player's movement direction, movement speed, movement time, movement trajectory, reaction time, posture, jump, serve, spike, and receive.

[0176] For example, movement data (180) may include indicators indicating the movement of the ball, such as the direction of movement, movement speed, movement time, movement trajectory, spin direction, and number of spins of the ball in the target volleyball game.

[0177] For example, the match situation data (170) may include indicators indicating the match situation of the target volleyball match, such as events that occurred in the target volleyball match (e.g., rally, deuce, rally point), rule violations by each team (e.g., caution, warning, ejection), set scores of each team, scores of each team per set, and referee decision results.

[0178] For example, in acquiring movement data (180), the processor (210) may receive image data (160) of a target volleyball game from the camera group (130). In this case, the processor (210) may generate frame-by-frame position data (125) of a player and frame-by-frame position data (125) of a ball in the target volleyball game based on an image set included in the image data (160). The processor (210) may generate movement data (180) of a player based on the frame-by-frame position data (125) of the player. The processor (210) may generate movement data (180) of a ball based on the frame-by-frame position data (125) of the ball.

[0179] For example, in obtaining game situation data (170), the processor (210) may receive game situation data (170) of a target volleyball game from an external device (140). Additionally or alternatively, the processor (210) may generate game situation data (170) based on image data (160) received from a camera group (130).

[0180] At step S1220, the processor (210) can obtain analysis data (190) indicating the presence or absence of an abnormality in the target volleyball game based on at least one of movement data (180) or game situation data (170).

[0181] For example, the processor (210) may input at least one of movement data (180) or game situation data (170) into a learning model (800) trained to detect the presence or absence of an abnormality in a volleyball game, and obtain analysis data (190) indicating the presence or absence of an abnormality in the target volleyball game as an output of the learning model (800).

[0182] For example, the learning model (800) may be a model learned using a first movement pattern of a player in one or more other volleyball matches as learning data (810). Here, the first movement pattern may refer to statistical data on indicators indicating the movement of the player, such as the movement direction, movement speed, movement time, movement trajectory, reaction time, posture, jump, serve, spike, and receive of the player in one or more other volleyball matches. The processor (210) may extract a second movement pattern of the player from the movement data (180). Here, the second movement pattern may be statistical data on indicators indicating the actual movement of the player, such as the movement direction, movement speed, movement time, movement trajectory, reaction time, posture, jump, serve, spike, and receive of the player in the target volleyball match. The processor (210) can obtain analysis data (190) from a second movement pattern using a learning model (800) that has learned a first movement pattern of a player in one or more other volleyball games. The processor (210) can input the second movement pattern as input data (820) into the learning model (800) and obtain analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) can generate expected movement data of the player in the target volleyball game based on the first movement pattern learned in advance. Here, the expected movement data can include indicators indicating expected movements of the corresponding player, such as a movement direction, a movement speed, a movement time, a movement trajectory, a reaction time, a posture, a jump, a serve, a spike, and a receive. The learning model (800) can generate analysis data (190) indicating the presence or absence of an abnormality in the target volleyball game based on the second movement pattern and the expected movement data.For example, the analysis data (190) may include at least one of a first movement pattern, a second movement pattern, a comparison result between the first movement pattern and the second movement pattern, expected movement data, or a comparison result between the second movement pattern and the expected movement data. In addition, the analysis data (190) may also include an indicator or notification indicating that an abnormal behavior of a specific player has been detected in the target volleyball match. Here, based on the comparison result between the first movement pattern and the second movement pattern or the comparison result between the second movement pattern and the expected movement data, it may be determined whether the analysis data (190) includes an indicator or notification indicating that an abnormal behavior of a specific player has been detected in the target volleyball match. As a result of comparing the first movement pattern and the second movement pattern, if the player's serve success rate or receive success rate decreases by a predetermined value or more, the processor (210) may determine to include in the analysis data (190) an indicator or notification indicating that the player is behaving abnormally in the target volleyball game. As a result of comparing the second movement pattern and the expected movement data, if the difference between the player's expected jump height (starting point) and the actual jump height (starting point) in a specific situation is a predetermined value or more, the processor (210) may determine to include in the analysis data (190) an indicator or notification indicating that the player is behaving abnormally in the target volleyball game.

[0183] For example, the learning model (800) may be a model trained using a first game situation pattern of another volleyball game as learning data (810). For example, the first game situation pattern may refer to statistical data on indicators indicating the game situation of the volleyball game, such as events that occurred in another volleyball game (e.g., deuce, rally point), rule violations of each team (e.g., caution, warning, red card), set scores of each team, scores of each team per set, and referee decision results. The processor (210) may input game situation data (170) of the target volleyball game into the learning model (800) to obtain analysis data (190) indicating the presence or absence of an abnormality in the target volleyball game as an output of the learning model (800). The processor (210) may extract a second game situation pattern of the target volleyball game based on the game situation data (170) of the target volleyball game. The second match situation pattern may refer to statistical data on indicators indicating the actual match situation of the target volleyball match, such as events that occurred in the target volleyball match (e.g., deuce, rally point), rule violations of each team (e.g., caution, warning, ejection), set scores of each team, scores of each team per set, and referee decision results. The processor (210) may obtain analysis data (190) from the second match situation pattern using a learning model (800) that has learned the first match situation pattern of another volleyball match. Specifically, the processor (210) may input the second match situation pattern as input data (820) into the learning model (800) and obtain the analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) may generate expected match situation data of the target volleyball match based on the first match situation pattern learned in advance.Here, the expected match situation data may include indicators indicating the expected match situation of the target volleyball game, such as events, rule violations of each team, scores of each team, game leading time of each team, and referee decisions. The learning model (800) may generate analysis data (190) indicating the presence or absence of an abnormality in the target volleyball game based on the second match situation pattern and the expected match situation data. For example, the analysis data (190) may include at least one of the first match situation pattern, the second match situation pattern, the comparison result between the first match situation pattern and the second match situation pattern, the expected match situation data, or the comparison result between the second match situation pattern and the expected match situation data. In addition, the analysis data (190) may include an indicator or notification indicating the presence of an abnormal match situation in the target volleyball game. Here, based on the comparison results between the first match situation pattern and the second match situation pattern or the comparison results between the second match situation pattern and the expected match situation data, it may be determined whether the analysis data (190) includes an indicator or notification indicating that an abnormal match situation exists in the target volleyball game. If the comparison results between the second match situation pattern and the expected match situation data indicate that the referee's expected decision and the actual decision for a specific situation differ, the processor (210) may determine to include an indicator or notification indicating that an abnormal match situation exists in the analysis data (190).

[0184] For example, the learning model (800) may be a model learned using the first game situation pattern of each of a plurality of different volleyball games as learning data (810). The processor (210) may input the game situation data (170) of the target volleyball game into the learning model (800) to obtain analysis data (190) indicating the presence or absence of an abnormality in the target volleyball game as an output of the learning model (800). The processor (210) may extract the second game situation pattern of the target volleyball game based on the game situation data (170) of the target volleyball game. The processor (210) may obtain analysis data (190) from the second game situation pattern using the learning model (800) that has learned the first game situation pattern of each of a plurality of different volleyball games. Specifically, the processor (210) can input the second game situation pattern as input data (820) into the learning model (800) and obtain the analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) can generate predicted game situation data of the target volleyball game based on the first game situation patterns of each of a plurality of other volleyball games learned in advance. Here, the predicted game situation data can include indicators indicating the predicted game situation of the target volleyball game, such as win / loss prediction and score prediction. The learning model (800) can generate analysis data (190) indicating the presence or absence of an abnormality in the target volleyball game based on the second game situation pattern and the predicted game situation data. For example, the analysis data (190) may include at least one of a first game situation pattern, a second game situation pattern of each of a plurality of different volleyball games, a comparison result between the first game situation pattern and the second game situation pattern of each of a plurality of different volleyball games, expected game situation data, or a comparison result between the second game situation pattern and the expected game situation data.Additionally, the analysis data (190) may include an indicator or notification indicating that an abnormal game situation exists in the target volleyball game. Here, based on the comparison results between the first game situation pattern and the second game situation pattern of each of a plurality of other volleyball games or the comparison results between the second game situation pattern and the expected game situation data, it may be determined whether the analysis data (190) includes an indicator or notification indicating that an abnormal game situation exists in the target volleyball game. If the comparison results between the first game situation pattern of each of a plurality of other volleyball games and the second game situation pattern of the target volleyball game indicate that a specific team loses to an opposing team by a specific set score in the target volleyball game, and the number of games in which the specific team loses by the set score among the plurality of other volleyball games is greater than or equal to a predetermined value, the processor (210) may determine to include in the analysis data (190) an indicator or notification indicating that an abnormal game situation exists by a specific team in the target volleyball game.

[0185] Below, an example of a baseball game is described.

[0186] The processor (210) can perform the method (1200) to detect the presence or absence of an abnormality in a baseball game.

[0187] In step S1210, the processor (210) may obtain at least one of movement data (180) regarding the movement of an object (120) in a target baseball game or game situation data (170) indicating a game situation in the target baseball game. In the target baseball game, the object (120) may be a pitcher, a batter, a catcher, an infielder, an outfielder, a base runner, an umpire, a ball, etc. For convenience of explanation, the explanation will be centered on the pitcher.

[0188] For example, movement data (180) may include indicators indicating the movement of the pitcher in the target baseball game, such as the pitching posture and pitching time.

[0189] For example, movement data (180) may include indicators indicating the movement of the ball in the target baseball game, such as the direction of movement, movement speed, movement time, movement trajectory, spin direction, and number of spins.

[0190] For example, the game situation data (170) may include indicators indicating the game situation of the target baseball game, such as pitchers, batters, runners, ball / strike / out counts, scores, and umpire decisions.

[0191] For example, in acquiring movement data (180), the processor (210) may receive image data (160) of a target baseball game from the camera group (130). In this case, the processor (210) may generate frame-by-frame position data (125) of a pitcher and frame-by-frame position data (125) of a ball in the target baseball game based on an image set included in the image data (160). The processor (210) may generate movement data (180) of a pitcher based on the frame-by-frame position data (125) of the pitcher. The processor (210) may generate movement data (180) of a ball based on the frame-by-frame position data (125) of the ball.

[0192] For example, in obtaining game situation data (170), the processor (210) may receive game situation data (170) of a target baseball game from an external device (140). Additionally or alternatively, the processor (210) may generate game situation data (170) based on image data (160) received from a camera group (130).

[0193] At step S1220, the processor (210) can obtain analysis data (190) indicating the presence or absence of an abnormality in the target baseball game based on at least one of movement data (180) or game situation data (170).

[0194] For example, the processor (210) may input at least one of movement data (180) or game situation data (170) into a learning model (800) trained to detect the presence or absence of an abnormality in a baseball game, and obtain analysis data (190) indicating the presence or absence of an abnormality in the target baseball game as an output of the learning model (800).

[0195] For example, the learning model (800) may be a model learned using a pitcher's first movement pattern in one or more other baseball games as learning data (810). Here, the first movement pattern may refer to statistical data on indicators indicating the pitcher's movement, such as the pitching posture and pitching time of the pitcher in one or more other baseball games. The processor (210) may extract a pitcher's second movement pattern from the movement data (180). Here, the second movement pattern may be statistical data on indicators indicating the actual movement of the pitcher, such as the pitching posture and pitching time of the pitcher in the target baseball game. The processor (210) may obtain analysis data (190) from the second movement pattern using the learning model (800) that learned the pitcher's first movement pattern in one or more other baseball games. The processor (210) can input the second movement pattern as input data (820) into the learning model (800) and obtain analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) can generate predicted movement data of the pitcher in the target baseball game based on the first movement pattern learned in advance. Here, the predicted movement data can include indicators indicating the predicted movement of the corresponding pitcher, such as the pitching posture and the time required for pitching. The learning model (800) can generate analysis data (190) indicating the presence or absence of an abnormality in the target baseball game based on the second movement pattern and the predicted movement data. For example, the analysis data (190) can include at least one of the first movement pattern, the second movement pattern, the comparison result between the first movement pattern and the second movement pattern, the predicted movement data, or the comparison result between the second movement pattern and the predicted movement data.Additionally, the analysis data (190) may include an indicator or notification indicating that abnormal behavior of a specific pitcher has been detected in the target baseball game. Here, based on the comparison result between the first movement pattern and the second movement pattern or the comparison result between the second movement pattern and the expected movement data, it may be determined whether the analysis data (190) includes an indicator or notification indicating that abnormal behavior of a specific pitcher has been detected in the target baseball game. If the comparison result between the first movement pattern and the second movement pattern indicates that a change in the pitcher's pitching posture (e.g., the angle of the throwing arm) is outside a predetermined range, the processor (210) may determine to include an indicator or notification indicating that abnormal behavior of the pitcher has been detected in the analysis data (190).

[0196] For example, the learning model (800) may be a model learned using a first game situation pattern of another baseball game as learning data (810). For example, the first game situation pattern may refer to statistical data on indicators indicating the game situation of the corresponding baseball game, such as pitch distribution, hitting distribution, ball / strike / out count, score, and umpire decision results in another baseball game. The processor (210) may input game situation data (170) of the target baseball game into the learning model (800) to obtain analysis data (190) indicating the presence or absence of an abnormality in the target baseball game as an output of the learning model (800). The processor (210) may extract a second game situation pattern of the target baseball game based on the game situation data (170) of the target baseball game. The second game situation pattern may refer to statistical data on indicators indicating the actual game situation of the target baseball game, such as pitch distribution, hitting distribution, ball / strike / out count, score, and umpire decision results in the target baseball game. The processor (210) can obtain analysis data (190) from a second game situation pattern using a learning model (800) that has learned a first game situation pattern of another baseball game. Specifically, the processor (210) can input the second game situation pattern as input data (820) into the learning model (800) and obtain analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) can generate expected game situation data of a target baseball game based on the first game situation pattern learned in advance. Here, the expected game situation data can include indicators indicating the expected game situation of the target baseball game, such as pitch distribution, hitting distribution, ball / strike / out count, score, and umpire decision results.The learning model (800) can generate analysis data (190) indicating whether there is an abnormality in the target baseball game based on the second game situation pattern and the expected game situation data. For example, the analysis data (190) can include at least one of the first game situation pattern, the second game situation pattern, the comparison result between the first game situation pattern and the second game situation pattern, the expected game situation data, or the comparison result between the second game situation pattern and the expected game situation data. In addition, the analysis data (190) can include an indicator or notification indicating that there is an abnormal game situation in the target baseball game. Here, whether the analysis data (190) includes an indicator or notification indicating that there is an abnormal game situation in the target baseball game can be determined based on the comparison result between the first game situation pattern and the second game situation pattern or the comparison result between the second game situation pattern and the expected game situation data. If, as a result of comparing the second game situation pattern and the expected game situation data, the umpire's expected decision on whether a specific pitch is a ball / strike and the actual decision are different, the processor (210) may decide to include an indicator or notification in the analysis data (190) indicating that there is an abnormal game situation in the target baseball game.

[0197] For example, the learning model (800) may be a model learned using the first game situation pattern of each of a plurality of different baseball games as learning data (810). The processor (210) may input the game situation data (170) of the target baseball game into the learning model (800) to obtain analysis data (190) indicating the presence or absence of an abnormality in the target baseball game as an output of the learning model (800). The processor (210) may extract the second game situation pattern of the target baseball game based on the game situation data (170) of the target baseball game. The processor (210) may obtain analysis data (190) from the second game situation pattern using the learning model (800) that has learned the first game situation pattern of each of a plurality of different baseball games. Specifically, the processor (210) can input the second game situation pattern as input data (820) into the learning model (800) and obtain the analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) can generate expected game situation data of the target baseball game based on the first game situation patterns of each of a plurality of other baseball games learned in advance. Here, the expected game situation data can include indicators indicating the expected game situation of the target baseball game, such as pitch distribution, hitting distribution, ball / strike / out count, score, and umpire decision results. The learning model (800) can generate analysis data (190) indicating the presence or absence of an abnormality in the target baseball game based on the second game situation pattern and the expected game situation data. For example, the analysis data (190) may include at least one of a first game situation pattern, a second game situation pattern of each of a plurality of different baseball games, a comparison result between the first game situation pattern and the second game situation pattern of each of a plurality of different baseball games, expected game situation data, or a comparison result between the second game situation pattern and the expected game situation data.Additionally, the analysis data (190) may include an indicator or notification indicating that an abnormal game situation exists in the target baseball game. Here, whether the analysis data (190) includes an indicator or notification indicating that an abnormal game situation exists in the target baseball game may be determined based on a comparison result between the first game situation pattern and the second game situation pattern of each of a plurality of different baseball games, or a comparison result between the second game situation pattern and expected game situation data.

[0198] Below, an example of a table tennis match is described.

[0199] The processor (210) can perform the method (1200) to detect the presence or absence of an abnormality in a table tennis match.

[0200] In step S1210, the processor (210) may obtain at least one of movement data (180) regarding the movement of an object (120) in a target table tennis match or match situation data (170) indicating a match situation in the target table tennis match. In the target table tennis match, the object (120) may be a player and a ball.

[0201] For example, movement data (180) may include indicators indicating the movement of the player in the target table tennis match, such as the player's movement direction, movement speed, movement time, movement trajectory, reaction time, posture, jump, serve, smash, grip, etc.

[0202] For example, movement data (180) may include indicators indicating the movement of the ball in the target table tennis match, such as the direction of movement, movement speed, movement time, movement trajectory, spin direction, and number of spins.

[0203] For example, the match situation data (170) may include indicators indicating the match situation of the target table tennis match, such as events that occurred in the target table tennis match (e.g., rally), rule violations of each team, set scores of each player (team), scores of each player (team) per set, and referee decision results.

[0204] For example, in acquiring movement data (180), the processor (210) may receive image data (160) of a target table tennis match from the camera group (130). In this case, the processor (210) may generate frame-by-frame position data (125) of a player and frame-by-frame position data (125) of a ball in the target table tennis match based on an image set included in the image data (160). The processor (210) may generate movement data (180) of a player based on the frame-by-frame position data (125) of the player. The processor (210) may generate movement data (180) of a ball based on the frame-by-frame position data (125) of the ball.

[0205] For example, in obtaining match situation data (170), the processor (210) may receive match situation data (170) of a target table tennis match from an external device (140). Additionally or alternatively, the processor (210) may generate match situation data (170) based on image data (160) received from the camera group (130).

[0206] At step S1220, the processor (210) can obtain analysis data (190) indicating the presence or absence of an abnormality in the target table tennis match based on at least one of movement data (180) or match situation data (170).

[0207] For example, the processor (210) may input at least one of movement data (180) or match situation data (170) into a learning model (800) trained to detect the presence or absence of an abnormality in a table tennis match, and obtain analysis data (190) indicating the presence or absence of an abnormality in the target table tennis match as an output of the learning model (800).

[0208] For example, the learning model (800) may be a model learned using a first movement pattern of a player in one or more other table tennis matches as learning data (810). Here, the first movement pattern may refer to statistical data on indicators indicating the movement of the player, such as the movement direction, movement speed, movement time, movement trajectory, reaction time, posture, jump, serve, smash, grip, etc. in one or more other table tennis matches. The processor (210) may extract a second movement pattern of the player from the movement data (180). Here, the second movement pattern may be statistical data on indicators indicating the actual movement of the player, such as the movement direction, movement speed, movement time, movement trajectory, reaction time, posture, jump, serve, smash, grip, etc. in the target table tennis match. The processor (210) can obtain analysis data (190) from a second movement pattern using a learning model (800) that has learned a first movement pattern of a player in one or more other table tennis matches. The processor (210) can input the second movement pattern as input data (820) into the learning model (800) and obtain analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) can generate expected movement data of the player in the target table tennis match based on the first movement pattern learned in advance. Here, the expected movement data can include indicators indicating the expected movement of the corresponding player, such as a movement direction, a movement speed, a movement time, a movement trajectory, a reaction time, a posture, a jump, a serve, a smash, and a grip. The learning model (800) can generate analysis data (190) indicating the presence or absence of an abnormality in the target table tennis match based on the second movement pattern and expected movement data.For example, the analysis data (190) may include at least one of a first movement pattern, a second movement pattern, a comparison result between the first movement pattern and the second movement pattern, expected movement data, or a comparison result between the second movement pattern and the expected movement data. In addition, the analysis data (190) may also include an indicator or notification indicating that an abnormal behavior of a specific player has been detected in the target table tennis match. Here, whether the analysis data (190) includes an indicator or notification indicating that an abnormal behavior of a specific player has been detected in the target table tennis match may be determined based on the comparison result between the first movement pattern and the second movement pattern or the comparison result between the second movement pattern and the expected movement data. As a result of comparing the second movement pattern and the expected movement data, if the difference between the expected movement trajectory and the actual movement trajectory of the player on the court in a specific situation is outside a predetermined range, the processor (210) may decide to include an indicator or notification in the analysis data (190) indicating that there is an abnormal behavior of the player in the target table tennis match.

[0209] For example, the learning model (800) may be a model learned using a first match situation pattern of another table tennis match as learning data (810). For example, the first match situation pattern may refer to statistical data on indicators indicating the match situation of the corresponding table tennis match, such as events that occurred in another table tennis match (e.g., rally), rule violations of each team, set scores of each player (team), scores of each player (team) per set, and referee decision results. The processor (210) may input match situation data (170) of the target table tennis match into the learning model (800) to obtain analysis data (190) indicating the presence or absence of an abnormality in the target table tennis match as an output of the learning model (800). The processor (210) may extract a second match situation pattern of the target table tennis match based on the match situation data (170) of the target table tennis match. The second match situation pattern may refer to statistical data on indicators indicating the actual match situation of the target table tennis match, such as events that occurred in the target table tennis match (e.g., rallies), rule violations of each team, set scores of each player (team), scores of each player (team) per set, and referee decision results. The processor (210) may obtain analysis data (190) from the second match situation pattern using a learning model (800) that has learned the first match situation pattern of another table tennis match. Specifically, the processor (210) may input the second match situation pattern as input data (820) into the learning model (800) and obtain the analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) may generate expected match situation data of the target table tennis match based on the first match situation pattern learned in advance. Here, the expected match situation data may include indicators indicating the expected match situation of the target table tennis match, such as an event that occurred (e.g., a rally), a rule violation by each team, a set score of each player (team), a score of each player (team) per set, and a referee's decision result.The learning model (800) can generate analysis data (190) indicating whether there is an abnormality in the target table tennis match based on the second match situation pattern and the expected match situation data. For example, the analysis data (190) can include at least one of the first match situation pattern, the second match situation pattern, the comparison result between the first match situation pattern and the second match situation pattern, the expected match situation data, or the comparison result between the second match situation pattern and the expected match situation data. In addition, the analysis data (190) can include an indicator or notification indicating that there is an abnormal match situation in the target table tennis match. Here, whether the analysis data (190) includes an indicator or notification indicating that there is an abnormal match situation in the target table tennis match can be determined based on the comparison result between the first match situation pattern and the second match situation pattern or the comparison result between the second match situation pattern and the expected match situation data.

[0210] For example, the learning model (800) may be a model learned using the first match situation pattern of each of a plurality of different table tennis matches as learning data (810). The processor (210) may input match situation data (170) of the target table tennis match into the learning model (800) to obtain analysis data (190) indicating the presence or absence of an abnormality in the target table tennis match as an output of the learning model (800). The processor (210) may extract the second match situation pattern of the target table tennis match based on the match situation data (170) of the target table tennis match. The processor (210) may obtain analysis data (190) from the second match situation pattern using the learning model (800) that has learned the first match situation pattern of each of a plurality of different table tennis matches. Specifically, the processor (210) can input the second match situation pattern as input data (820) into the learning model (800) and obtain the analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) can generate expected match situation data of the target table tennis match based on the first match situation patterns of each of a plurality of other table tennis matches learned in advance. Here, the expected match situation data can include indicators indicating the expected match situation of the target table tennis match, such as an event that occurred (e.g., a rally), a rule violation of each team, a set score of each player (team), a score of each player (team) per set, and a referee's decision result. The learning model (800) can generate analysis data (190) indicating the presence or absence of an abnormality in the target table tennis match based on the second match situation pattern and the expected match situation data. For example, the analysis data (190) may include at least one of a first match situation pattern, a second match situation pattern of each of a plurality of different table tennis matches, a comparison result between the first match situation pattern and the second match situation pattern of each of a plurality of different table tennis matches, expected match situation data, or a comparison result between the second match situation pattern and the expected match situation data.Additionally, the analysis data (190) may include an indicator or notification indicating that an abnormal match situation exists in the target table tennis match. Here, based on the comparison results between the first match situation pattern and the second match situation pattern of each of a plurality of other table tennis matches or the comparison results between the second match situation pattern and the expected match situation data, it may be determined whether the analysis data (190) includes an indicator or notification indicating that an abnormal match situation exists in the target table tennis match. If the comparison results between the first match situation pattern of each of a plurality of other table tennis matches and the second match situation pattern of the target table tennis match indicate that a specific player in the target table tennis match loses to an opposing player by a specific set score, and the number of matches in which the specific player loses by the specific set score among the plurality of other table tennis matches is outside a predetermined range, the processor (210) may determine to include in the analysis data (190) an indicator or notification indicating that an abnormal match situation exists by a specific player in the target table tennis match.

[0211] Below, an example of a tennis match is described.

[0212] The processor (210) can perform the method (1200) to detect the presence or absence of an abnormality in a tennis match.

[0213] In step S1210, the processor (210) may obtain at least one of movement data (180) regarding the movement of an object (120) in a target tennis match or match situation data (170) indicating a match situation in the target tennis match. In the target tennis match, the object (120) may be a player and a ball.

[0214] For example, movement data (180) may include indicators indicating the movement of the player in the target tennis match, such as the player's movement direction, movement speed, movement time, movement trajectory, reaction time, posture, jump, serve, smash, etc.

[0215] For example, movement data (180) may include indicators indicating the movement of the ball in the target tennis match, such as the direction of movement, movement speed, movement time, movement trajectory, spin direction, and number of spins.

[0216] For example, the match situation data (170) may include indicators indicating the match situation of the target tennis match, such as events that occurred in the target tennis match (e.g., rallies), rule violations of each team, set scores of each player (team), scores of each player (team) per set, and referee decision results.

[0217] For example, in acquiring movement data (180), the processor (210) may receive image data (160) of a target tennis match from the camera group (130). In this case, the processor (210) may generate frame-by-frame position data (125) of a player and frame-by-frame position data (125) of a ball in the target tennis match based on an image set included in the image data (160). The processor (210) may generate movement data (180) of a player based on the frame-by-frame position data (125) of the player. The processor (210) may generate movement data (180) of a ball based on the frame-by-frame position data (125) of the ball.

[0218] For example, in obtaining match situation data (170), the processor (210) may receive match situation data (170) of a target tennis match from an external device (140). Additionally or alternatively, the processor (210) may generate match situation data (170) based on image data (160) received from the camera group (130).

[0219] At step S1220, the processor (210) can obtain analysis data (190) indicating the presence or absence of an abnormality in the target tennis match based on at least one of movement data (180) or match situation data (170).

[0220] For example, the processor (210) may input at least one of movement data (180) or match situation data (170) into a learning model (800) trained to detect the presence or absence of an abnormality in a tennis match, and obtain analysis data (190) indicating the presence or absence of an abnormality in the target tennis match as an output of the learning model (800).

[0221] For example, the learning model (800) may be a model learned using a first movement pattern of a player in one or more other tennis matches as learning data (810). Here, the first movement pattern may refer to statistical data on indicators indicating the movement of the player, such as the movement direction, movement speed, movement time, movement trajectory, reaction time, posture, jump, serve, smash, etc. in one or more other tennis matches. The processor (210) may extract a second movement pattern of the player from the movement data (180). Here, the second movement pattern may be statistical data on indicators indicating the actual movement of the player, such as the movement direction, movement speed, movement time, movement trajectory, reaction time, posture, jump, serve, smash, etc. in the target tennis match. The processor (210) may obtain analysis data (190) from the second movement pattern using the learning model (800) that learned the first movement pattern of the player in one or more other tennis matches. The processor (210) can input the second movement pattern as input data (820) into the learning model (800) and obtain analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) can generate expected movement data of the player in the target tennis match based on the first movement pattern learned in advance. Here, the expected movement data can include indicators indicating the expected movement of the player, such as movement direction, movement speed, movement time, movement trajectory, reaction time, posture, jump, serve, smash, etc. The learning model (800) can generate analysis data (190) indicating the presence or absence of an abnormality in the target tennis match based on the second movement pattern and the expected movement data.For example, the analysis data (190) may include at least one of a first movement pattern, a second movement pattern, a comparison result between the first movement pattern and the second movement pattern, expected movement data, or a comparison result between the second movement pattern and the expected movement data. In addition, the analysis data (190) may also include an indicator or notification indicating that abnormal behavior of a specific player has been detected in the target tennis match. Here, based on the comparison result between the first movement pattern and the second movement pattern or the comparison result between the second movement pattern and the expected movement data, it may be determined whether the analysis data (190) includes an indicator or notification indicating that abnormal behavior of a specific player has been detected in the target tennis match. If the comparison result between the first movement pattern and the second movement pattern indicates that the serve speed of the player has decreased beyond a predetermined range, the processor (210) may determine to include an indicator or notification indicating that abnormal behavior of the player has been detected in the analysis data (190). As a result of comparing the second movement pattern and the expected movement data, if the difference between the expected movement trajectory of the player and the actual movement trajectory in a specific situation is outside a predetermined range, the processor (210) may decide to include an indicator or notification in the analysis data (190) indicating that there is abnormal behavior by the player in the target tennis match.

[0222] For example, the learning model (800) may be a model trained using a first match situation pattern of another tennis match as training data (810). For example, the first match situation pattern may refer to statistical data on indicators indicating the match situation of the tennis match, such as events that occurred in another tennis match (e.g., a rally), rule violations of each team, set scores of each player (team), scores of each player (team) per set, and referee decision results. The processor (210) may input match situation data (170) of the target tennis match into the learning model (800) to obtain analysis data (190) indicating the presence or absence of an abnormality in the target tennis match as an output of the learning model (800). The processor (210) may extract a second match situation pattern of the target tennis match based on the match situation data (170) of the target tennis match. The second match situation pattern may refer to statistical data on indicators indicating the actual match situation of the target tennis match, such as events that occurred in the target tennis match (e.g., rallies), rule violations of each team, set scores of each player (team), scores of each player (team) per set, and referee decision results. The processor (210) may obtain analysis data (190) from the second match situation pattern using a learning model (800) that has learned the first match situation pattern of another tennis match.

[0223] Specifically, the processor (210) can input the second match situation pattern as input data (820) into the learning model (800) and obtain the analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) can generate expected match situation data of the target tennis match based on the first match situation pattern learned in advance. Here, the expected match situation data can include indicators indicating the expected match situation of the target tennis match, such as an event that occurred (e.g., a rally), a rule violation of each team, a set score of each player (team), a score of each player (team) per set, and a referee's decision result. The learning model (800) can generate analysis data (190) indicating the presence or absence of an abnormality in the target tennis match based on the second match situation pattern and the expected match situation data. For example, the analysis data (190) may include at least one of a first match situation pattern, a second match situation pattern, a comparison result between the first match situation pattern and the second match situation pattern, expected match situation data, or a comparison result between the second match situation pattern and the expected match situation data. In addition, the analysis data (190) may include an indicator or notification indicating that an abnormal match situation exists in the target tennis match. Here, based on the comparison result between the first match situation pattern and the second match situation pattern or the comparison result between the second match situation pattern and the expected match situation data, it may be determined whether the analysis data (190) includes an indicator or notification indicating that an abnormal match situation exists in the target tennis match. If the comparison result between the first match situation pattern and the second match situation pattern shows that the number of points conceded by a specific player (team) in a specific situation (e.g., a break point in a set) is greater than a predetermined range, the processor (210) may determine to include an indicator or notification indicating that an abnormal match situation exists in the analysis data (190).If, as a result of comparing the second match situation pattern and the expected match situation data, the referee's expected decision for a specific situation differs from the actual decision, the processor (210) may decide to include an indicator or notification in the analysis data (190) indicating that there is an abnormal match situation in the target tennis match.

[0224] For example, the learning model (800) may be a model learned using the first match situation pattern of each of a plurality of different tennis matches as learning data (810). The processor (210) may input match situation data (170) of the target tennis match into the learning model (800) to obtain analysis data (190) indicating the presence or absence of an abnormality in the target tennis match as an output of the learning model (800). The processor (210) may extract the second match situation pattern of the target tennis match based on the match situation data (170) of the target tennis match. The processor (210) may obtain analysis data (190) from the second match situation pattern using the learning model (800) that has learned the first match situation pattern of each of a plurality of different tennis matches. Specifically, the processor (210) can input the second match situation pattern as input data (820) into the learning model (800) and obtain the analysis data (190) as output data (830) of the learning model (800). In this case, the learning model (800) can generate expected match situation data of the target tennis match based on the first match situation patterns of each of a plurality of other tennis matches learned in advance. Here, the expected match situation data can include indicators indicating the expected match situation of the target tennis match, such as an event that occurred (e.g., a rally), a rule violation of each team, a set score of each player (team), a score of each player (team) per set, and a referee's decision result. The learning model (800) can generate analysis data (190) indicating the presence or absence of an abnormality in the target tennis match based on the second match situation pattern and the expected match situation data.For example, the analysis data (190) may include at least one of a first match situation pattern, a second match situation pattern of each of a plurality of different tennis matches, a comparison result between the first match situation pattern and the second match situation pattern of each of the plurality of different tennis matches, expected match situation data, or a comparison result between the second match situation pattern and the expected match situation data. In addition, the analysis data (190) may include an indicator or notification indicating that an abnormal match situation exists in the target tennis match. Here, whether the analysis data (190) includes an indicator or notification indicating that an abnormal match situation exists in the target tennis match may be determined based on the comparison result between the first match situation pattern and the second match situation pattern of each of the plurality of different tennis matches or the comparison result between the second match situation pattern and the expected match situation data. As a result of comparing the first match situation pattern of each of a plurality of other tennis matches with the second match situation pattern of the target tennis match, if a specific player in the target tennis match loses to an opposing player by a specific set score, and the number of matches among the plurality of other tennis matches in which the specific player loses by the set score is greater than a predetermined range, the processor (210) may determine to include in the analysis data (190) an indicator or notification indicating that there is an abnormal match situation by the specific player in the target tennis match.

[0225] The methods according to the present disclosure may be implemented using a device having a computer or processor. While the steps of the methods are illustrated and described in a predetermined order in this disclosure, the steps may be performed in any order that can be arbitrarily combined according to the present disclosure, in addition to being performed sequentially. In one embodiment, at least some of the steps may be performed in parallel, iteratively, or heuristically. The present disclosure does not exclude variations or modifications to the methods. In one embodiment, at least some of the steps may be omitted, or other steps may be added.

[0226] Various embodiments of the present disclosure may be implemented as software recorded on a machine-readable recording medium. The software may be software for implementing the various embodiments of the present disclosure described above. The software may be inferred from various embodiments of the present disclosure by programmers skilled in the art to which the present disclosure pertains. For example, the software may be machine-readable instructions (e.g., code or code segments) or a program. The device may be a device capable of operating according to instructions called from a recording medium, such as a computer. In one embodiment, the device may be a device (100) according to embodiments of the present disclosure. In one embodiment, the processor of the device may execute the called instructions, causing components of the device to perform functions corresponding to the instructions. In one embodiment, the processor may be a processor (210) according to embodiments of the present disclosure. The recording medium may refer to a recording medium on which data is stored and readable by the device. The recording medium may include, for example, ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc. In one embodiment, the recording medium may be memory (220). In one embodiment, the recording medium may also be implemented in a distributed form, such as in a network-connected computer system. Software may be distributed, stored, and executed in a computer system, etc. The recording medium may be a non-transitory recording medium. A non-transitory recording medium means a tangible medium regardless of whether data is stored semi-permanently or temporarily, and does not include a signal that is propagated transitorily.

[0227] Although the technical concept of the present disclosure has been described through various embodiments, the technical concept of the present disclosure encompasses various substitutions, modifications, and alterations that can be made within the scope understandable to those of ordinary skill in the art to which the present disclosure pertains. Furthermore, it should be understood that such substitutions, modifications, and alterations are included within the scope of the appended claims. Embodiments according to the present disclosure can be combined with each other. Each embodiment can be combined in various ways depending on the number of cases, and embodiments created by combining them also fall within the scope of the present disclosure.

Claims

1. Communication circuit; one or more processors; and comprising one or more memories storing instructions executed by the one or more processors; One or more of the above processors, Obtaining at least one of movement data regarding the movement of an object in a target sports game or game situation data indicating a game situation of the target sports game, Based on at least one of the movement data or the game situation data, it is configured to detect the presence or absence of an anomaly in the target sports game and obtain analysis data indicating the presence or absence of the anomaly. The one or more processors, in obtaining the movement data, Receiving image data for the target sports event from a camera group including one or more cameras through the above communication circuit, A device configured to generate the movement data based on the image data.

2. In paragraph 1, The one or more processors, in obtaining the game situation data, A device configured to receive game situation data indicating the game situation of the target sports game from another device through the communication circuit.

3. In paragraph 1, The one or more processors, in obtaining the game situation data, Receive the image data from the camera group through the communication circuit, A device configured to generate the game situation data based on the image data.

4. In paragraph 1, One or more of the above processors, A device configured to transmit at least one of the movement data, the game situation data, or the analysis data to a user terminal.

5. In paragraph 1, One or more of the above processors, A device configured to input at least one of the movement data or the game situation data into a learning model trained to detect the presence or absence of an abnormality in a sports game, and to obtain the analysis data indicating the presence or absence of an abnormality in the target sports game as an output of the learning model.

6. In paragraph 5, The above learning model is a model learned using the first movement pattern of the object in one or more other sports games as learning data, One or more of the above processors, Based on the movement data, a second movement pattern of the object is extracted in the target sports game, A device configured to obtain the analysis data from the second movement pattern using the learning model.

7. In paragraph 5, The above learning model is a model trained using the first game situation pattern of other sports games as learning data, One or more of the above processors, Based on the above match situation data, a second match situation pattern of the target sports game is extracted, A device configured to obtain the analysis data from the second game situation pattern using the above learning model.

8. In paragraph 5, The above learning model is a model learned using the first game situation pattern of each of multiple different sports games as learning data, One or more of the above processors, Based on the above match situation data, a second match situation pattern of the target sports game is extracted, A device configured to obtain the analysis data from the second game situation pattern using the above learning model.

9. In paragraph 1, A device wherein the image data comprises an image set divided into frames by capturing images of the target sports event from one or more angles by the camera group.

10. In paragraph 9, One or more of the above processors, Based on the above image set, it is configured to generate frame-by-frame location data of the object of the target sports game, A device wherein the frame-by-frame position data includes three-dimensional coordinates where the object is located.

11. In paragraph 10, One or more of the above processors, Based on the frame-by-frame position data, the movement data of the object is configured to be generated, A device wherein the movement data includes at least one of a movement direction, a movement speed, a movement time, or a movement trajectory of the object.

12. In paragraph 1, A device wherein the object comprises at least one of a player or a ball of the target sporting event.

13. In paragraph 1, A device wherein the above game situation data includes at least one of the game progress time, score, number of rule violations, game lead time, or referee decision result in the target sports game.

14. A method performed in a device including one or more processors and one or more memories storing instructions to be executed by the one or more processors, One or more of the above processors, A step of acquiring at least one of movement data regarding the movement of an object in a target sports game or game situation data indicating a game situation of the target sports game; and A step of detecting whether an anomaly exists in the target sports game based on at least one of the movement data or the game situation data, and obtaining analysis data indicating whether the anomaly exists, The step of acquiring the movement data comprises: A step of receiving image data for the target sporting event from a camera group including one or more cameras; and A method comprising a step of generating the movement data based on the image data.

15. In a non-transitory computer-readable recording medium recording instructions to be executed by one or more processors, The above instructions, when executing the above instructions, cause the one or more processors to: Obtaining at least one of movement data regarding the movement of an object in a target sports game or game situation data indicating the game situation of the target sports game, Based on at least one of the movement data or the game situation data, detecting the presence or absence of an anomaly in the target sports game and obtaining analysis data indicating the presence or absence of the anomaly, In obtaining the above movement data, the one or more processors, Receiving image data for the target sporting event from a camera group including one or more cameras, A non-transitory computer-readable recording medium that generates the movement data based on the image data.

Images