[go: up one dir, main page]

WO2018122957A1 - Système, procédé et programme de support d'analyse de mouvement sportif - Google Patents

Système, procédé et programme de support d'analyse de mouvement sportif Download PDF

Info

Publication number
WO2018122957A1
WO2018122957A1 PCT/JP2016/088885 JP2016088885W WO2018122957A1 WO 2018122957 A1 WO2018122957 A1 WO 2018122957A1 JP 2016088885 W JP2016088885 W JP 2016088885W WO 2018122957 A1 WO2018122957 A1 WO 2018122957A1
Authority
WO
WIPO (PCT)
Prior art keywords
image data
data
unit
sports
event
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/JP2016/088885
Other languages
English (en)
Japanese (ja)
Inventor
洋介 本橋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
NEC Corp
Original Assignee
NEC Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by NEC Corp filed Critical NEC Corp
Priority to PCT/JP2016/088885 priority Critical patent/WO2018122957A1/fr
Priority to JP2018558561A priority patent/JP6677320B2/ja
Publication of WO2018122957A1 publication Critical patent/WO2018122957A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B69/00—Training appliances or apparatus for special sports
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion

Definitions

  • the present invention relates to a sports motion analysis support system, a sports motion analysis support method, and a sports motion analysis support program that support motion analysis in sports.
  • Motion capture is known as a technique that can be used for motion analysis in sports.
  • application software that displays a trajectory of a specific part of the body (for example, an ankle) based on a moving image obtained by imaging a person who is playing sports.
  • application software that displays the difference between a model form and the form of a sports person.
  • Patent Document 1 discloses a system in which a skeleton image serving as a model and a skeleton image that matches a learner's body shape is superimposed and displayed on a video obtained by photographing a form such as a golf swing of a learner with a video camera. Are listed.
  • the following shows an example where the form and the result of the action including the form do not always correspond completely. For example, even if a person's form is good, it may happen that the person's condition was not good and the result was not good. Also, for example, even if a person's form is good, it may occur that good results are not achieved due to external factors such as rain and wind. Similarly, even if a person moves in a form in which the ball tends to fly in the right direction, the ball may fly in the left direction due to conditions or external factors.
  • users can present a video that represents a behavior that includes a form that is likely to produce a predetermined result statistically from many videos, the user can improve his / her own form or the competitor's form.
  • the video can be used to discover spiders.
  • Patent Document 1 displays a skeleton image superimposed on one moving image. Accordingly, when there are a large number of moving images, the user must check each moving image on which the skeleton images are superimposed one by one.
  • the present invention provides a sports motion analysis support system and a sports motion analysis support that can easily identify video image data representing a motion including a form in which a predetermined result is likely to occur from a lot of motion image data. It is an object to provide a method and a sports motion analysis support program.
  • a sports motion analysis support system includes a data storage unit that stores a plurality of pieces of data in which video image data representing a series of motions in sports and the results of the motions are associated with each other, and motions and results corresponding to the motions
  • a learning unit that learns a model representing the relationship between the image data, a prediction unit that calculates a predicted value of the result based on the model for each image data, and a data storage unit that is based on the predicted value calculated for each image data
  • a selection unit that selects a predetermined number of image data from the plurality of image data stored in the image data.
  • the sports motion analysis support method includes a computer including a data storage unit that stores a plurality of data in which image data of a moving image representing a series of motions in sports and the results of the motions are associated with each other.
  • the model representing the relationship with the result corresponding to the operation is learned, and the predicted value of the result is calculated based on the model for each image data, and the data storage unit is calculated based on the predicted value calculated for each image data.
  • a predetermined number of image data is selected from a plurality of stored image data.
  • a sports motion analysis support program is a sport mounted on a computer including a data storage unit that stores a plurality of data in which image data of a moving image representing a series of motions in sports and the results of the motions are associated with each other.
  • a motion analysis support program in which a computer learns a model representing a relationship between a motion and a result corresponding to the motion, and a prediction processing that calculates a predicted value based on the model for each image data And a selection process for selecting a predetermined number of pieces of image data from a plurality of pieces of image data stored in the data storage unit based on a predicted value calculated for each piece of image data.
  • FIG. 1 is a block diagram illustrating an example of a sports motion analysis support system according to a first embodiment of the present invention. It is a schematic diagram which shows a series of operation
  • Embodiment 1 a case will be described in which the result of a sport operation is represented by a numerical value indicating a score.
  • a sport a long jump will be described as an example. Therefore, in the present embodiment, a case where the result of the long jump operation is a numerical value indicating a score (in this example, a jump distance) will be described as an example.
  • the first embodiment can also be applied to other sports in which the result of the motion is represented by a numerical value indicating the result.
  • FIG. 1 is a block diagram showing an example of a sports motion analysis support system according to the first embodiment of the present invention.
  • the sports motion analysis support system 1 according to the first embodiment of the present invention includes a data storage unit 2, a learning unit 4, a prediction unit 5, an operation unit 8, a selection unit 6, and a display unit 7.
  • the data storage unit 2 is a storage device that stores a plurality of data in which image data of a moving image representing a series of actions in sports and the results of the actions are associated with each other.
  • image data of a moving image representing a series of actions in sports and the results of the actions are associated with each other.
  • moving image data representing a series of actions of a person who performs a long jump is associated with a result (jumping distance) obtained as a result of the action.
  • the data storage unit 2 stores a plurality of such data.
  • the manner of associating the image data with the results is not particularly limited.
  • the grade may exist as data different from the image data, and the image data and the grade may be associated with each other.
  • the image data and the results may be associated with each other in a manner in which the grade is included in the moving image of the image data (in other words, the manner in which the grade is expressed on the moving image). This also applies to embodiments described later.
  • the former will be described as an example.
  • FIG. 2 is a schematic diagram showing a series of actions of a person who performs a long jump (hereinafter referred to as a player).
  • a player a person who performs a long jump
  • FIG. 2 to simplify the drawing, for example, the same posture is illustrated as the posture of the player when running, but the actual player moves a series of limbs while moving the limbs and the like. Perform the action.
  • the athlete makes a run, jumps at the crossing board 11, and then lands.
  • moving image data representing the series of operations can be obtained.
  • the data which matched the image data and the result at that time become one data.
  • the data storage unit 2 stores a plurality of such data in advance.
  • the grade associated with the image data is an actual measurement value.
  • FIG. 3 is a schematic diagram showing an example of a plurality of data in which image data and results (actual measurement values) are associated with each other.
  • individual image data is indicated by a symbol in which a number “#” and a number are connected for convenience.
  • the data storage unit 2 stores, for example, data illustrated in FIG. 3 in advance. Although FIG. 3 shows 20 pieces of data, the number of data stored in the data storage unit 2 is not limited to 20, and more data may be stored.
  • a plurality of storage units 2 may be stored.
  • the data storage unit 2 stores a plurality of data related to a specific player.
  • the specific player may be a player who uses the sports motion analysis support system 1 of the present invention, or a player who is instructed by a coach using the sports motion analysis support system 1.
  • the specific player may be a player who is a competitor for the user who uses the sports motion analysis support system 1 of the present invention. This also applies to other embodiments described later.
  • the movie represents a series of actions of a player
  • the movie also represents the form of the player.
  • the learning unit 4 uses a plurality of data stored in the data storage unit 2 (see, for example, FIG. 3) as teacher data, and the motion of the player represented by the video and the results (jumping distance) corresponding to the motion.
  • the learning unit 4 When learning the model, the learning unit 4 extracts a still image from each image data of the moving image every predetermined time (for example, every 0.5 seconds). That is, the learning unit 4 extracts a set of still images from moving image data. This set of still images represents the movement and form of the player.
  • 0.5 second was illustrated as an example of the above-mentioned predetermined time here, the above-mentioned predetermined time is not limited to 0.5 second.
  • the learning unit 4 learns, as a model, a model in which a set of still images is an explanatory variable and a score is an objective variable. Therefore, it is possible to calculate the predicted value of the score using the model obtained by learning and the moving image data (more specifically, a set of still images extracted from the moving image data). .
  • the model learned by the learning unit 4 can also be said to be a model representing the relationship between the player's form and the grade.
  • the machine learning algorithm may be any algorithm that can learn a model for calculating a predicted value of results using moving image data.
  • the prediction unit 5 calculates the predicted value of the results for each image data stored in the data storage unit 2 based on the model learned by the learning unit 4.
  • the prediction unit 5 may extract a set of still images from the image data when calculating the predicted value of the grade for one image data. And the prediction part 5 should just calculate the predicted value of a grade by applying the set of the still image to a model as an explanatory variable.
  • the selection unit 6 corresponds to each image data in which the predicted value of the score calculated by the predicting unit 5 corresponds from the upper first to the upper predetermined number, and the predicted value of the result corresponds to the lower first to the lower predetermined number Each image data being selected is selected.
  • the values representing the upper predetermined number and the lower predetermined number are specified by the user of the sports motion analysis support system 1. It can be said that the values representing the upper predetermined number and the lower predetermined number are selection criteria when the selection unit 6 selects image data.
  • the operation unit 8 is a user interface for the user to input values representing the upper predetermined number and the lower predetermined number to the sports motion analysis support system 1 (more specifically, the selection unit 6).
  • the operation unit 8 is a user interface for the user to input selection criteria.
  • the operation unit 8 is realized by an input device such as a keyboard, for example.
  • the mode of the operation unit 8 is not limited to such an input device.
  • the selection unit 6 corresponds to each image data in which the predicted value of the score corresponds from the top first to the top predetermined number, and the predicted value of the score corresponds from the bottom to the bottom predetermined number.
  • the case where each image data being selected is selected has been described.
  • the selection unit 6 selects each image data corresponding to the predicted value of the grade from the first highest to the upper predetermined number, and for each image data corresponding to the predicted value from the lower first to the lower predetermined number May not be selected.
  • the selection unit 6 selects each image data corresponding to the predicted value of the grade from the lower first to the lower predetermined number, and each image whose predicted value corresponds to the upper first to the upper predetermined number There is no need to select data.
  • each image data in which the selection unit 6 corresponds to the predicted value of the grade from the top first to the top predetermined number, and the image in which the predicted value of the grade corresponds to the bottom first to the lower predetermined number A case where data is selected will be described as an example.
  • each piece of image data corresponding to the predicted value from the top first to the top predetermined is image data of a moving image predicted to have good results.
  • each piece of image data corresponding to the predicted value from the lower first to the lower predetermined number is image data of a moving image that is predicted to have poor results.
  • each piece of image data having a predicted value corresponding to the top first to the top predetermined is video image data that is predicted to have poor results. Further, each piece of image data corresponding to the predicted values from the lower first to the lower predetermined number is image data of a moving image predicted to have good results.
  • the selection unit 6 selects the image data corresponding to the predicted value of the grade from the top 1 to the top number, and the predicted value of the grade corresponds to the lowest 1 to the lowest order. It may be determined in advance whether the selected image data is selected. That is, the selection criterion may be determined in advance. In that case, the operation unit 8 may not be provided. The selection unit 6 may be set in advance so as to select the image data with the highest grade predicted value and the image data with the lowest grade predicted value.
  • the display unit 7 displays a moving image of each image data selected by the selection unit 6.
  • the display unit 7 includes each image data corresponding to the predicted value of the grade from the first highest to the highest predetermined number, and each image data corresponding to the predicted value of the grade from the lower first to the lower predetermined number Are displayed separately.
  • the display unit 7 displays, on the right side of the screen, the moving image of each image data corresponding to the predicted value of the grade from the top first to the top predetermined number, and the grade predicted value is from the bottom first to the lower predetermined number.
  • You may display the moving image of each image data applicable to the left of the screen.
  • the mode of displaying both separately is not limited to the above example.
  • Learning unit 4, prediction unit 5, selection unit 6 and display unit 7 are realized by, for example, a CPU (Central Processing Unit) of a computer having a display device (not shown in FIG. 1).
  • the CPU reads a sports motion analysis support program from a program recording medium such as a computer program storage device (not shown in FIG. 1), and in accordance with the sports motion analysis support program, the learning unit 4, the prediction unit 5, and the selection unit 6 and the display unit 7 may be operated.
  • the display unit 7 a part that actually displays a moving image is realized by a display device, and a part that performs display control of a moving image based on image data is realized by a CPU.
  • the computer may be a personal computer or a portable computer such as a smartphone. These points are the same in other embodiments described later.
  • the sports motion analysis support system 1 may have a configuration in which two or more physically separated devices are connected by wire or wirelessly.
  • the sports motion analysis support system 1 may be realized as a system in which a portable computer such as a smartphone and a server cooperate. This also applies to other embodiments described later.
  • FIG. 4 is a flowchart showing an example of processing progress of the first embodiment of the present invention.
  • description may be abbreviate
  • the data storage unit 2 stores a plurality of data in advance.
  • a description will be given assuming that a plurality of data shown in FIG.
  • the learning unit 4 uses the data stored in the data storage unit 2 (see FIG. 3) as teacher data to learn a model that represents the relationship between the motion of the player represented by the video and the grade corresponding to the motion. (Step S1). In step S ⁇ b> 1, the learning unit 4 extracts a still image from the individual image data of the moving image every predetermined time (for example, every 0.5 seconds). As a result, the set of still images is associated with the grade. Then, the learning unit 4 only needs to learn a model in which a set of still images is used as an explanatory variable and a score is used as an objective variable. As described above, this model can also be said to be a model representing the relationship between the player's form and the grade.
  • the prediction unit 5 calculates a predicted value of the grade based on the model obtained in step S1 for each image data included in the teacher data (see FIG. 3) (step S2). As described above, the prediction unit 5 extracts a set of still images from the image data and calculates the set of still images as an explanatory variable to the model when calculating the predicted value of the grade for one image data. By doing so, the predicted value of the grade may be calculated. The prediction unit 5 may perform this process for each image data.
  • FIG. 5 is an explanatory diagram showing an example of the predicted value obtained in step S2. In this example, it is assumed that the predicted value of the result shown in FIG. 5 is calculated for each image data in step S2.
  • the selection unit 6 receives an input of selection criteria from the user via the operation unit 8 (step S3).
  • the selection unit 6 receives input of values representing the upper predetermined number and the lower predetermined number via the operation unit 8.
  • “2” is input as a value representing the upper predetermined number and the lower predetermined number will be described as an example.
  • the selection unit 6 selects image data based on the predicted value of the score calculated in step S2 and the selection criterion input in step S3 (step S4).
  • the selection unit 6 corresponds to each image data in which the predicted value of the grade corresponds to the top first to the top second, and the predicted value of the grade corresponds to the lowest first to the lowest second. Select each image data.
  • the top first predicted value is “8m95”, and this predicted value corresponds to image data # 5.
  • the second highest predicted value is “8m93”, and this predicted value corresponds to the image data # 4.
  • the lower first prediction value is “7m00”, and this prediction value corresponds to the image data # 17.
  • the lower second of the predicted value is “7m50”, and this predicted value corresponds to the image data # 14.
  • the selection unit 6 selects # 5 and # 4 as the respective image data corresponding to the predicted values from the upper first to the upper second, and the predicted values are from the lower first to the lower 2 # 17 and # 14 are selected as the image data corresponding to the first.
  • the display unit 7 displays a moving image of the image data selected in step S4 (step S5).
  • the display unit 7 displays the moving images # 5 and # 4 and the moving images # 17 and # 14, respectively.
  • the model is generated using a plurality of combinations of moving images and results. Therefore, when the predicted value of the grade is calculated by this model, it can be said that the predicted value statistically well represents the tendency of the grade according to the form.
  • the selection unit 6 uses the image data corresponding to the predicted values from the top first to the upper predetermined position and the image data corresponding to the predicted values from the upper first to the upper predetermined position.
  • the display unit 7 displays a moving image of each image data.
  • the video of each image data corresponding to the predicted value from the top first to the top predetermined number (assumed to be the nth) has a score of n from the video that is statistically predicted to have the best grade.
  • the moving image of each image data corresponding to the predicted value from the lower first to the lower predetermined number (assumed to be the nth) is the nth from the video that is statistically predicted to have the worst result.
  • factors that do not appear in the form are excluded, and the video when the results are predicted to be statistically good and the results are statistically It is possible to display a video that is predicted to be bad. By checking these videos, the user can expect a statistically good form or a statistically bad result without considering conditions or external factors. You can check the form and analyze the form.
  • the image data when the actual result is the best is # 4
  • the image data indicating the form in which the best result is statistically best is # 5.
  • the moving image of the image data # 4 is also displayed, but the user performs analysis for improving the form, for example, by analyzing the image of the image data # 5 more particularly. Can do.
  • # 17 shown in FIG. 5 is a video when the actual result was the third worst.
  • the predicted value of the score of # 17 is the first lower order. Therefore, it can be said that the moving image # 17 represents a form in which statistically worst results are likely to be obtained.
  • the user can perform analysis for improving the form, for example, by confirming the moving image # 17 selected by the selection unit 6.
  • video image data representing an operation including a form that easily produces good results or an image of a movie representing an operation including a form that easily produces bad results.
  • Data can be easily identified.
  • the selection criteria may be determined in advance.
  • a selection criterion that prescribes selection of the first image data with the highest predicted value and the first image data with the lower predicted value of the grade may be determined in advance.
  • the learning unit 4 uses a model that uses external factors and player conditions as explanatory variables. You may learn.
  • the prediction unit 5 may substitute 0 for the explanatory variable of the external factor in the model, or substitute a value representing “good” for the explanatory variable of the condition. .
  • the predicted value of the result excluding the external factors and the change of the condition is obtained, and as a result, the same effect as described above can be obtained.
  • external factors and condition data are required during model learning.
  • the plurality of data storage units 2 may store data in which moving image data, results, external factors, and conditions are associated with each other.
  • An example of such data is shown in FIG.
  • the learning unit 4 can learn a model in which not only a moving image but also a wind speed and a condition of a tail wind are explanatory variables using a plurality of data illustrated in FIG. 6 as teacher data.
  • “0” means “good” and “1” means “bad”.
  • Embodiment 2 a case will be described in which the result of a sports operation is represented by an event.
  • a PK penalty kick
  • the result of the PK action is one of two types of events: “the ball flew to the right” and “the ball flew to the left”.
  • the second embodiment can be applied to other sports in which the result of the operation is represented as an event.
  • the sports motion analysis support system according to the second embodiment of the present invention can be represented by the block diagram shown in FIG. 1 similarly to the sports motion analysis support system 1 according to the first embodiment.
  • a second embodiment will be described. Explanation of matters similar to those in the first embodiment will be omitted as appropriate.
  • a sports motion analysis support system 1 according to the second embodiment includes a data storage unit 2, a learning unit 4, a prediction unit 5, an operation unit 8, a selection unit 6, and a display unit 7 (see FIG. 1). ).
  • the data storage unit 2 is a storage device that stores a plurality of data in which image data of a moving image representing a series of actions in sports and the results of the actions are associated with each other.
  • image data of a moving image representing a series of actions in sports and the results of the actions are associated with each other.
  • moving image image data representing a series of actions of a person who performs PK hereinafter referred to as a player
  • an event (“the ball flew to the right” as a result of the actions).
  • the data storage unit 2 stores a plurality of such data.
  • “right” is assumed to be “right” when viewed from the player who performed PK. The same applies to “left”.
  • the result (event) of the PK operation is represented by binary values.
  • FIG. 7 is a schematic diagram showing a series of actions of a player performing PK.
  • the same posture is illustrated as the posture of the player performing PK, but the actual player performs a series of operations while moving the limbs and the like.
  • the player who performs PK makes a run toward the ball 21 placed in front of the goalkeeper 22 and kicks the ball 21. Then, the ball 21 flies to the right or left. That is, event “1” or event “0” occurs.
  • By capturing a series of actions of a player performing PK with a video camera moving image data representing the series of actions can be obtained. And the data which matched the image data and the event at that time ("1" or "0”) become one data.
  • the data storage unit 2 stores a plurality of such data in advance.
  • FIG. 8 is a schematic diagram showing an example of a plurality of data in which image data and events are associated with each other. Similarly to FIG. 3, each image data is indicated by a symbol in which a number “#” and a number are connected for convenience. Although 20 pieces of data are shown in FIG. 8, the number of pieces of data stored in the data storage unit 2 is not limited to 20, and more pieces of data may be stored.
  • the learning unit 4 uses a plurality of data (for example, see FIG. 8) stored in the data storage unit 2 as teacher data, and represents the relationship between the motion of the player represented by the video and the result corresponding to the motion.
  • the model is learned by machine learning. More specifically, the learning unit 4 learns a model that represents the relationship between the motion of the player and the probability that the event “1” will occur or the probability that the event “0” will occur.
  • the learning unit 4 When learning the model, the learning unit 4 extracts a still image from each image data of the moving image every predetermined time (for example, every 0.2 seconds). That is, the learning unit 4 extracts a set of still images from moving image data. That is, the learning unit 4 extracts a set of still images from moving image data. This operation is the same as the operation of the learning unit 4 in the first embodiment. As described in the first embodiment, this set of still images represents the movement and form of the player.
  • 0.2 second was illustrated as an example of the above-mentioned predetermined time here, the above-mentioned predetermined time is not limited to 0.2 second.
  • the learning unit 4 learns, as a model, a model in which a set of still images is used as an explanatory variable, and a probability that an event “1” occurs or a probability that an event “0” occurs is an objective variable. Therefore, the probability of event “1” or event “0” using a model obtained by learning and moving image data (more specifically, a set of still images extracted from moving image data). Can be calculated.
  • a case where the probability that the event “1” occurs or the probability that the event “0” occurs is represented by one objective variable that can take a value in the range of 0 to 1 will be described as an example.
  • the value of the objective variable represents the probability that the event “1” (that is, the event “the ball flies to the right”) will occur. The closer it is, the higher the probability that the event “1” will occur. The closer the value is to 0.5, the lower the probability that the event “1” will occur. If the value of the objective variable is smaller than 0.5, the value represents the probability that the event “0” (that is, the event “the ball flies to the left”) will occur, and the value is zero. The closer it is, the higher the probability that an event “0” will occur. The closer the value is to 0.5, the lower the probability that an event “0” will occur.
  • the learning unit 4 may use, for example, logistic regression analysis as a machine learning algorithm.
  • the learning unit 4 may learn a model representing the relationship between the player's motion and the probability of occurrence of the event “1” or the probability of occurrence of the event “0” by logistic regression analysis.
  • the model learned by the learning unit 4 can also be said to be a model representing the relationship between the player's form and the probability of occurrence of event “1” or the probability of occurrence of event “0”.
  • the learning unit 4 may learn a model in which the probability that the event “1” occurs and the probability that the event “0” occurs are objective variables.
  • the probability that the event “1” occurs is, for example, a value in the range of 0 to 1, and the higher the value, the higher the probability that the event “1” occurs.
  • the probability that the event “0” occurs is, for example, a value in the range of 0 to 1, and the larger the value, the higher the probability that the event “0” occurs.
  • the sum of the probability of occurrence of event “1” and the probability of occurrence of event “0” may be 1.
  • the value of the objective variable is greater than 0.5
  • the value represents the probability of occurrence of event “1”
  • the value of the objective variable is less than 0.5, the probability of occurrence of event “0”.
  • the probability is low.
  • the value is smaller than 0.5, the closer the value is to 0, the higher the probability that an event “0” will occur. The closer the value is to 0.5, the lower the probability that an event “0” will occur. Represents.
  • the prediction unit 5 calculates the probability that the event “1” occurs or the probability that the event “0” occurs for each image data stored in the data storage unit 2 based on the model learned by the learning unit 4.
  • the probability that the event “1” will occur or the probability that the event “0” will occur can be said to be one of the aspects of the predicted value of the result corresponding to the action.
  • the prediction unit 5 may extract a set of still images from the image data when calculating the probability for one image data. Then, the prediction unit 5 may calculate the probability (the value of the objective variable) by applying the set of still images as an explanatory variable to the model.
  • values representing events actually generated by the action are, for example, names It can be said that it is a scale.
  • the value of the objective variable calculated by the prediction unit 5 based on the model represents the probability.
  • the possible values of the objective variable are continuous values in the range of 0 to 1. Therefore, it can be said that the value of the objective variable calculated by the prediction unit 5 based on the model is, for example, an order scale.
  • the selection unit 6 corresponds to each image data in which the value of the objective variable calculated by the prediction unit 5 corresponds to the upper first to the upper predetermined number, and the value of the objective variable corresponds to the lower first to the lower predetermined number.
  • Each image data being selected is selected.
  • Values representing the upper predetermined number and the lower predetermined number are specified by the user of the sports motion analysis support system 1. This value can be said to be a selection criterion when the selection unit 6 selects image data.
  • the operation unit 8 is the same as the operation unit 8 in the first embodiment, and a description thereof will be omitted.
  • the selection unit 6 corresponds to each image data in which the value of the objective variable corresponds to the upper first to the upper predetermined number, and the value of the objective variable corresponds to the lower first to the lower predetermined number.
  • the case where each image data being selected is selected has been described.
  • the selection unit 6 selects each image data in which the value of the objective variable calculated by the prediction unit 5 corresponds to the upper first to the upper predetermined number, and the value of the objective variable is from the lower first to the lower predetermined number. It is not necessary to select each image data corresponding to.
  • the selection unit 6 selects each image data in which the value of the objective variable calculated by the prediction unit 5 corresponds to the first lower order to the lower predetermined order, and the value of the objective variable is the upper first to the upper predetermined. It is not necessary to select each image data corresponding to the first.
  • the selection unit 6 selects each image data in which the value of the objective variable corresponds to the upper first to the upper predetermined position, and each image in which the value of the objective variable corresponds to the lower first to the lower predetermined position.
  • a case where data is selected will be described as an example.
  • each piece of image data corresponding to the value of the objective variable from the top first to the top predetermined is image data of a moving image that is predicted to have a high probability of occurrence of the event “1”.
  • each piece of image data in which the value of the objective variable corresponds from the lower first to the predetermined lower order is image data of a moving image that is predicted to have a high probability of occurrence of the event “0”.
  • the selection criteria may be determined in advance. In that case, the operation unit 8 may not be provided.
  • the selection unit 6 may be set in advance so as to select the image data having the highest value of the objective variable and the image data having the lowest value of the objective variable.
  • the display unit 7 displays a moving image of each image data selected by the selection unit 6.
  • the display unit 7 displays each image data corresponding to the value of the objective variable from the upper first to the upper predetermined position, and each image data corresponding to the value of the objective variable from the lower first to the lower predetermined position. Distinguish and display.
  • the display unit 7 may display the moving image of the former image data on the right side of the screen and display the latter image data on the left side of the screen.
  • the mode of displaying both separately is not limited to the above example.
  • FIG. 9 is a flowchart showing an example of processing progress of the second embodiment of the present invention.
  • description may be abbreviate
  • the data storage unit 2 stores a plurality of data in advance.
  • a description will be given assuming that a plurality of data shown in FIG.
  • the learning unit 4 uses the data stored in the data storage unit 2 (see FIG. 8) as teacher data, and the motion of the player represented by the video and the probability that the event “1” will occur or the event “0”. A model representing the relationship with the probability of occurrence of learning is learned (step S11).
  • the learning unit 4 may learn a model by logistic regression analysis, for example.
  • step S11 the learning unit 4 extracts still images from the individual image data of the moving image every predetermined time.
  • the set of still images is associated with the event “1” or “0” that actually occurred.
  • the learning unit 4 may learn a model in which a set of still images is used as an explanatory variable, and a probability that an event “1” occurs or a probability that an event “0” occurs is an objective variable.
  • this model can also be said to be a model representing the relationship between the player's form and the probability of occurrence of event “1” or the probability of occurrence of event “0”.
  • the prediction unit 5 calculates the value of the objective variable for each image data included in the teacher data (see FIG. 8) based on the model obtained in step S11 (step S12).
  • This objective variable represents the probability that the event “1” will occur or the probability that the event “0” will occur.
  • the prediction unit 5 extracts a set of still images from the image data, and uses the set of still images as an explanatory variable. What is necessary is just to calculate the value of an objective variable by applying to a model.
  • the prediction unit 5 may perform this process for each image data.
  • FIG. 10 is an explanatory diagram showing an example of the probability (the value of the objective variable) obtained in step S12. In the following description, it is assumed that the probability (value of the objective variable) shown in FIG. 10 is calculated for each image data in step S12.
  • the selection unit 6 receives an input of selection criteria from the user via the operation unit 8 (step S13).
  • the selection unit 6 receives input of values representing the upper predetermined number and the lower predetermined number via the operation unit 8.
  • “2” is input as a value representing the upper predetermined number and the lower predetermined number will be described as an example.
  • the selection unit 6 selects image data based on the value of the objective variable calculated in step S12 and the selection criterion input in step S13 (step S14).
  • the selection unit 6 corresponds to each image data in which the value of the objective variable corresponds to the upper first to the upper second, and the value of the objective variable corresponds to the lower first to the lower second. Select each image data.
  • the selection unit 6 selects # 13 and # 1 as the respective image data corresponding to the value of the objective variable from the upper first to the upper second, and the value of the objective variable is from the lower first.
  • # 4 and # 7 are selected as the image data corresponding to the lower second (see FIG. 10).
  • the display unit 7 displays the moving image of the image data selected in step S14 (step S15).
  • the display unit 7 displays the moving images # 13 and # 1, and the moving images # 4 and # 7, respectively.
  • the model is generated using a plurality of combinations of moving images and events. Therefore, it can be said that the value calculated based on the model and representing the probability of occurrence of event “1” or the probability of occurrence of event “0” statistically represents the tendency corresponding to the form.
  • the learning unit 4 generates a model having the probability that the event “1” occurs or the probability that the event “0” occurs as an objective variable. That is, the event in the teacher data is not a continuous value, but the value of the objective variable calculated based on the model is a continuous value. Therefore, it is possible to easily rank image data based on the value of the objective variable. Therefore, even if the number of image data is large, the image data of a movie that is statistically predicted when the event “1” occurs (that is, the image data of the movie that is statistically predicted when the ball flies to the right), It is possible to easily identify moving image data that is statistically predicted when the event “0” occurs (that is, moving image data that is statistically predicted that the ball flies to the left).
  • the user can analyze the form when the ball is predicted to fly to the right and the form when the ball is predicted to be statistically predicted to fly to the left. .
  • the user of the sports motion analysis support system 1 is a player who performs PK.
  • the user stores the image data and event of his / her moving image in the data storage unit 2 as teacher data, for example, so as to find a wrinkle of his / her form and eliminate the wrinkle of the form. It can be used for improvement. If such a habit disappears, it becomes difficult for the goalkeeper of the opponent team to predict the direction of the ball from the form, which is advantageous to the player who performs the PK.
  • the user of the sports motion analysis support system 1 may be a goal keeper.
  • the user of the sports motion analysis support system 1 may be a goal keeper.
  • image data and events of a player's moving image in the data storage unit 2 as teacher data
  • analysis for the purpose of finding that player's habit can be performed. If the opponent team's trap is discovered, it becomes easier for the goalkeeper to predict the direction of the ball after the PK from the player's form of the opponent team, which is advantageous for the goalkeeper.
  • the soccer PK has been described as an example.
  • the data storing unit 2 stores data in which image data obtained by capturing a pitching motion of a baseball pitcher and an event are associated with each other. It may be memorized. In this case, for example, “the ball type was straight”, “the ball type was a curve”, “the ball type was a fork”, and the like may be events.
  • the learning part 4 should just produce
  • the selection unit 6 also includes, for example, an upper predetermined number of image data with a high probability that the ball type is straight, an upper predetermined number of image data with a high probability that the ball type is a curve, and a probability that the ball type is a fork.
  • the upper predetermined number of image data having a higher value may be selected.
  • the sports motion analysis support system 1 may include a data acquisition unit that acquires data to be stored in the data storage unit 2 from the outside.
  • FIG. 11 is a block diagram illustrating a configuration example in the case where a data acquisition unit is provided.
  • the data storage unit 2, the learning unit 4, the prediction unit 5, the operation unit 8, the selection unit 6, and the display unit 7 are the same as those elements in the first embodiment and those in the second embodiment. The description is omitted.
  • the data acquisition unit 9 acquires a plurality of pieces of data in which moving image data representing a series of motions in sports and the results of the motions are associated with each other, and stores them in the data storage unit 2.
  • the data acquisition unit 9 may access the device, acquire a plurality of data from the device, and store the data in the data storage unit 2.
  • the processing after the data acquisition unit 9 stores a plurality of data in the data storage unit 2 is the same as the processing described in the first embodiment or the processing described in the second embodiment.
  • the data acquisition unit 9 is realized by, for example, a CPU of a computer that operates according to a sports motion analysis support program.
  • the long jump, PK in soccer, pitcher pitching in baseball, etc. are exemplified.
  • the operation of the sport to which the present invention is applied is not limited to these.
  • the data storage unit 2 may store data associating image data of an action tossed by a volleyball setter with a result indicating whether the ball flew to the right or left.
  • the present invention can be applied to a rugby formation, an American football formation, and the like.
  • the present invention can be applied to various sports operations.
  • FIG. 12 is a schematic block diagram showing a configuration example of a computer according to each embodiment of the present invention.
  • the computer 1000 includes a CPU 1001, a main storage device 1002, an auxiliary storage device 1003, an interface 1004, a display device 1005, and an input device 1006.
  • the input device 1006 corresponds to the operation unit 8.
  • the sports motion analysis support system 1 is implemented in a computer 1000.
  • the operation of the sports motion analysis support system 1 is stored in the auxiliary storage device 1003 in the form of a program (sport motion analysis support program).
  • the CPU 1001 reads out the program from the auxiliary storage device 1003, develops it in the main storage device 1002, and executes the above processing according to the program.
  • the auxiliary storage device 1003 is an example of a tangible medium that is not temporary.
  • Other examples of the non-temporary tangible medium include a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, and a semiconductor memory connected via the interface 1004.
  • this program is distributed to the computer 1000 via a communication line, the computer 1000 that has received the distribution may develop the program in the main storage device 1002 and execute the above processing.
  • the program may be for realizing a part of the above-described processing.
  • the program may be a differential program that realizes the above-described processing in combination with another program already stored in the auxiliary storage device 1003.
  • circuitry IV circuitry IV
  • processors or combinations thereof. These may be configured by a single chip or may be configured by a plurality of chips connected via a bus. Part or all of each component may be realized by a combination of the above-described circuit and the like and a program.
  • the plurality of information processing devices and circuits may be centrally arranged or distributedly arranged.
  • the information processing apparatus, the circuit, and the like may be realized as a form in which each is connected via a communication network, such as a client and server system and a cloud computing system.
  • FIG. 13 is a block diagram showing an outline of the present invention.
  • the sports motion analysis support system of the present invention includes a data storage unit 2, a learning unit 4, a prediction unit 5, and a selection unit 6.
  • the data storage unit 2 stores a plurality of pieces of data in which moving image data representing a series of actions in sports and results of the actions are associated with each other.
  • the learning unit 4 learns a model representing the relationship between an action and a result corresponding to the action.
  • the prediction unit 5 calculates the predicted value of the result based on the model for each image data.
  • the selection unit 6 selects a predetermined number of image data from the plurality of image data stored in the data storage unit 2 based on the predicted value calculated for each image data.
  • the result of the operation associated with the image data may be a numerical value indicating the score
  • the learning unit 4 may learn a model that represents the relationship between the motion and the numerical value indicating the score.
  • the result of the action associated with the image data may be an event, and the learning unit 4 may learn a model that represents the relationship between the action and the probability that the event will occur.
  • the learning unit 4 may be configured to learn a model representing the relationship between the motion and the probability of occurrence of an event by logistic regression analysis.
  • the present invention can be suitably applied to a sports motion analysis support system that supports motion analysis in sports.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Physical Education & Sports Medicine (AREA)
  • Image Analysis (AREA)
  • Processing Or Creating Images (AREA)

Abstract

Une unité de stockage de données 2 stocke une pluralité d'éléments de données vidéo montrant une série de mouvements dans un sport, associés aux résultats de ces mouvements. Une unité d'apprentissage 4 apprend un modèle représentant la relation entre les mouvements et les résultats correspondant à ces mouvements. Une unité de prédiction 5 calcule une valeur prédite des résultats pour chaque élément de données d'image, sur la base du modèle. Une unité de sélection 6 sélectionne un nombre prédéterminé d'éléments de données d'image stockés dans l'unité de stockage de données 2, sur la base de la valeur prédite calculée pour chaque élément de données d'image.
PCT/JP2016/088885 2016-12-27 2016-12-27 Système, procédé et programme de support d'analyse de mouvement sportif Ceased WO2018122957A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/JP2016/088885 WO2018122957A1 (fr) 2016-12-27 2016-12-27 Système, procédé et programme de support d'analyse de mouvement sportif
JP2018558561A JP6677320B2 (ja) 2016-12-27 2016-12-27 スポーツ動作解析支援システム、方法およびプログラム

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2016/088885 WO2018122957A1 (fr) 2016-12-27 2016-12-27 Système, procédé et programme de support d'analyse de mouvement sportif

Publications (1)

Publication Number Publication Date
WO2018122957A1 true WO2018122957A1 (fr) 2018-07-05

Family

ID=62707960

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2016/088885 Ceased WO2018122957A1 (fr) 2016-12-27 2016-12-27 Système, procédé et programme de support d'analyse de mouvement sportif

Country Status (2)

Country Link
JP (1) JP6677320B2 (fr)
WO (1) WO2018122957A1 (fr)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022097196A1 (fr) * 2020-11-04 2022-05-12 日本電信電話株式会社 Dispositif de détermination de mouvement cible temporaire, procédé de détermination de mouvement cible temporaire et programme
JP2022102577A (ja) * 2020-12-25 2022-07-07 エヌ・ティ・ティ・コミュニケーションズ株式会社 評価装置、評価方法および評価プログラム
CN115068920A (zh) * 2022-07-12 2022-09-20 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) 运动训练方法、装置、设备及存储介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080312010A1 (en) * 2007-05-24 2008-12-18 Pillar Vision Corporation Stereoscopic image capture with performance outcome prediction in sporting environments
WO2015080063A1 (fr) * 2013-11-27 2015-06-04 株式会社ニコン Appareil électronique
JP2015119833A (ja) * 2013-12-24 2015-07-02 カシオ計算機株式会社 運動支援システム及び運動支援方法、運動支援プログラム

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080312010A1 (en) * 2007-05-24 2008-12-18 Pillar Vision Corporation Stereoscopic image capture with performance outcome prediction in sporting environments
WO2015080063A1 (fr) * 2013-11-27 2015-06-04 株式会社ニコン Appareil électronique
JP2015119833A (ja) * 2013-12-24 2015-07-02 カシオ計算機株式会社 運動支援システム及び運動支援方法、運動支援プログラム

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022097196A1 (fr) * 2020-11-04 2022-05-12 日本電信電話株式会社 Dispositif de détermination de mouvement cible temporaire, procédé de détermination de mouvement cible temporaire et programme
JP2022102577A (ja) * 2020-12-25 2022-07-07 エヌ・ティ・ティ・コミュニケーションズ株式会社 評価装置、評価方法および評価プログラム
CN115068920A (zh) * 2022-07-12 2022-09-20 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) 运动训练方法、装置、设备及存储介质
CN115068920B (zh) * 2022-07-12 2023-08-11 合肥综合性国家科学中心人工智能研究院(安徽省人工智能实验室) 运动训练方法、装置、设备及存储介质

Also Published As

Publication number Publication date
JPWO2018122957A1 (ja) 2019-03-28
JP6677320B2 (ja) 2020-04-08

Similar Documents

Publication Publication Date Title
US12138042B2 (en) Methods and systems for facilitating interactive training of body-eye coordination and reaction time
US12370429B2 (en) Computer vision and artificial intelligence applications for performance evaluation and/or skills development
JP6915701B2 (ja) 抽出プログラム、抽出方法および情報処理装置
US20160042652A1 (en) Body-motion assessment device, dance assessment device, karaoke device, and game device
RU2645553C2 (ru) Устройство и способ для автоматической оценки хода тренировочного упражнения
US20240428439A1 (en) User analytics using a camera device and associated systems and methods
JP6677319B2 (ja) スポーツ動作解析支援システム、方法およびプログラム
CN108694466A (zh) 击球路线的推荐方法及装置
US20210170230A1 (en) Systems and methods for training players in a sports contest using artificial intelligence
CN112204570B (zh) 动态地确定区域
Jiang et al. Golfpose: Golf swing analyses with a monocular camera based human pose estimation
JP6677320B2 (ja) スポーツ動作解析支援システム、方法およびプログラム
US20220143485A1 (en) Display method, non-transitory computer-readable storage medium, and information processing apparatus
JP6760610B2 (ja) 位置計測システム、及び位置計測方法
JP7622542B2 (ja) 動作改善装置、動作改善方法、動作改善プログラム及び動作改善システム
US20220035456A1 (en) Methods and systems for performing object detection and object/user interaction to assess user performance
JP7740693B2 (ja) 戦術分析装置およびその制御方法、並びに制御プログラム
JP2020054748A (ja) プレイ分析装置、及び、プレイ分析方法
JP2020054747A (ja) プレイ分析装置、及び、プレイ分析方法
KR102835721B1 (ko) 경기 영상 분석을 위한 텍스트 라벨을 제공하기 위한 장치, 방법 및 기록 매체
WO2020217844A1 (fr) Dispositif de traitement de données, procédé de traitement de données, et programme
WO2020071092A1 (fr) Dispositif d'analyse de jeu et procédé d'analyse de jeu
US20250010132A1 (en) Evaluation apparatus, evaluation method, and program
JP2021022153A (ja) データ処理装置、データ処理方法、およびプログラム
WO2023106201A1 (fr) Dispositif d'analyse de jeu, procédé d'analyse de jeu et support de stockage lisible par ordinateur

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16925118

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2018558561

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16925118

Country of ref document: EP

Kind code of ref document: A1