JP2019024550A - Detection device, detection system, processing device, detection method and detection program - Google Patents

Abstract

To provide instruction image data and the like that is suitable for a user.SOLUTION: A detection device includes: a detector for detecting a human body of a user; and an update part for updating teacher data using posture data of the user based on a comparison result acquired by comparing the posture data of the user that is acquired from a detection result by the detector to the teacher data of the user that was acquired in the past.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a detection device, a detection system, a processing device, a detection method, and a detection program.

  A technique for providing instructional video data to a person (user) in motion and a technique for capturing an image of the person in motion in an appropriate state have been proposed (for example, see Patent Document 1 below). However, such instructional video data may not be optimal data for the user.

JP 2007-288682 A

  According to the first aspect of the present invention, the detection unit for detecting the user's human body, the user posture data obtained from the detection result of the detection unit, and the user's teacher data acquired in the past are obtained. There is provided a detection device including an update unit that updates teacher data using user posture data based on the comparison result.

  According to the second aspect of the present invention, the detection unit for detecting the human body of the user, the comparison unit for comparing the posture data of the user obtained from the detection result of the detection unit, and the teacher data of the user acquired in the past And a processing unit that generates user teacher data using user posture data based on the comparison result of the comparison unit.

  According to the third aspect of the present invention, a detection system including the detection device according to the first aspect or the second aspect is provided.

  According to the fourth aspect of the present invention, based on the result of comparison between the posture data of the human body obtained from the detection result of detecting the human body of the user and the user's teacher data acquired in the past, the posture data is used. A processing device is provided that includes an updating unit that updates teacher data.

  According to the fifth aspect of the present invention, the posture data is detected based on the result of comparison between the detection of the human body of the user, the posture data of the human body obtained from the detection, and the teacher data of the user acquired in the past. Using the method to update the teacher data.

  According to the sixth aspect of the present invention, based on the comparison result between the posture data of the human body obtained from the detection result of detecting the human body of the user and the teacher data of the user acquired in the past, A detection program is provided that executes updating teacher data using.

It is a figure which shows the detection apparatus which concerns on 1st Embodiment. It is a figure which shows the process of the detection apparatus which concerns on 1st Embodiment. It is a figure which shows the process of the comparison part which concerns on 1st Embodiment. It is a flowchart which shows the detection method which concerns on embodiment. It is a figure which shows the detection apparatus which concerns on 2nd Embodiment. It is a figure which shows the process of the detection apparatus which concerns on 2nd Embodiment. It is a figure which shows the process of the comparison part which concerns on 2nd Embodiment. It is a figure which shows the detection apparatus which concerns on 3rd Embodiment. It is a figure which shows the detection apparatus which concerns on 4th Embodiment. It is a figure which shows the process of the comparison part which concerns on 4th Embodiment. It is a figure which shows the detection apparatus which concerns on 5th Embodiment. It is a figure which shows the detection apparatus which concerns on 6th Embodiment. It is a figure which shows the detection apparatus which concerns on 7th Embodiment. It is a figure which shows the detection apparatus which concerns on 8th Embodiment. It is a figure which shows the detection apparatus which concerns on embodiment.

[First Embodiment]
A first embodiment will be described. FIG. 1 is a diagram illustrating a detection device according to an embodiment. For example, the detection device 1 detects the human body HM of the user U who takes a posture so as to approach the target (basic) posture (pose, form). The detection device 1 is, for example, a variety of sports (eg, baseball, soccer, golf, gymnastics, exercise, yoga, bodybuilding), fashion shows (eg, walking, posing), games, or person authentication (eg, security). Available in the field. For example, the user U may be a player of a game with the theme of yoga or golf, and the detection device 1 may detect the human body HM of the user U who plays the game.

  The detection device 1 includes, for example, a posture detection device, a motion detection device, an exercise support device, a posture evaluation device, and the like. The detection system SYS according to the embodiment includes a detection device 1. The detection system SYS in the present embodiment may include a plurality of detection devices 1, or may include one or both of the detection device 1, a display unit 4 and an input unit 5 described later.

  In the following description, it is assumed that the detection device 1 detects a user U taking a predetermined posture in yoga (eg, crescent moon pose, cat pose, cow pose, warrior pose). In addition, when the detection apparatus 1 detects the human body HM, it may be in a state in which an attachment such as clothing, shoes, or a protector (for example, an object that moves with the human body HM) is attached. The detection of the human body HM by the detection device 1 may include the detection of an attachment that moves with the human body HM. Further, the detection apparatus 1 may detect an object placed in a target posture, or a living organism or a robot (eg, a humanoid robot) that takes the target posture.

  The detection device 1 compares posture data FD (eg, current posture data FD of the current human body HM) obtained from the current detection for the user U with the teacher data TD (executes comparison processing). The posture data FD is information representing the posture (eg, two-dimensional shape, three-dimensional shape) of the human body HM of the user U. In the present embodiment, the posture data FD includes posture data related to the user's yoga. For example, in FIG. 1, the posture data FD includes information indicating the posture of the user U taking a yoga crescent pose. The posture data FD includes, for example, information representing at least a part of the outline (eg, silhouette) of the human body HM. The posture data FD includes, for example, information indicating the elongation or bending of each joint of the human body HM. Further, the posture data FD includes, for example, information that displays the joints (joint points) in the human body HM that are estimated and identified from the detection results (eg, images) of the detection unit 2 by connecting with lines.

  In addition, the posture data FD may include information that can specify the positional relationship of each part of the human body. For example, the posture data FD may include position information (eg, absolute position, relative position, coordinates) of a part of the human body HM. The position information of the human body includes, for example, one or both of the position of the end portion (eg, hand, foot, head) of the human body HM and the position of the joint (eg, shoulder, elbow, wrist, knee) of the human body HM. Including. The position information of the human body HM includes, for example, the position between the end portion and the joint (eg, neck, forearm, lower limb) and the position between the joint and joint (eg, upper arm, upper limb). And one or both of them.

  The teacher data TD is posture data (hereinafter referred to as past posture data) obtained from a detection result of detection performed on the same user U before the detection of the current (current) user U by the detection device 1. including. For example, the teacher data TD is, for example, posture data obtained from a detection result obtained by the detection apparatus 1 detecting the human body HM before this time, or posture data set at an initial stage (including the time of shipment). The teacher data TD may be posture data obtained from a detection result in which a detection device different from the detection device 1 detects the human body HM before this time.

  The detection device 1 updates and sets the teacher data TD using the posture data FD of the user U based on the comparison result of the comparison process (executes the update process). For example, the user U may cause the detection device 1 to perform an update process according to progress (eg, level) of yoga. For example, the user U may take into account his / her physique, physique, flexibility, etc. (eg, weighting and the like are factorized), and cause the detection apparatus 1 to set the teacher data TD. For example, the user U may perform yoga with reference to the updated teacher data TD as new teacher data. After performing yoga, the user U may select, for example, posture data determined to be suitable for himself as teacher data.

  The detection device 1 (or the detection system SYS) includes a detection unit 2, a processing device 3, a display unit 4, and an input unit 5. The detection unit 2 optically detects the human body HM of the user U. The processing device 3 processes information obtained from the detection result of the detection unit 2 (eg, a light reception result such as a captured image). The processing device 3 compares the posture data FD of the user U and the teacher data TD of the user U. The processing device 3 sets and updates the teacher data TD using the posture data FD based on the comparison result between the posture data FD and the teacher data TD. The detection device 1 may not include one or both of the display unit 4 and the input unit 5. For example, one or both of the display unit 4 (eg, display device) and the input unit 5 (input device) are external devices of the detection device 1, and the detection system SYS may include the external device described above. . Hereinafter, each part of the detection apparatus 1 will be described.

  The detection unit 2 includes, for example, a sensor 11 and a storage unit 12. The sensor 11 detects light from the human body HM of the user U. For example, the detection unit 2 irradiates the human body HM of the user U with illumination light, and the sensor 11 detects light emitted from the human body HM due to irradiation of the illumination light. The illumination light may include visible light or may include invisible light (eg, infrared light, ultraviolet light). Further, the light detected by the sensor 11 may include visible light, or may include non-visible light (eg, infrared light, ultraviolet light).

  For example, the sensor 11 performs a detection process at a predetermined frequency (eg, sampling rate, frame rate). The sensor 11 includes a light receiving element such as an imaging element, for example. The image sensor is, for example, a CCD image sensor or a CMOS image sensor. For example, the sensor 11 optically detects the human body HM by an imaging process. As the detection result, the sensor 11 outputs, for example, data of a captured image (hereinafter referred to as a captured image) (eg, pixel values such as RGB data and gray scale).

  The sensor 11 detects the human body HM from a predetermined position (arrangement position) and direction (detection direction) with respect to the human body HM. For example, the human body HM performs yoga in a predetermined region (for example, an exercise region such as a yoga mat or an exercise mat), and the sensor 11 is previously positioned with respect to the predetermined region.

  For example, before the user U performs yoga, the detection unit 2 may image the predetermined area using the sensor 11. The detection apparatus 1 may display an image captured by the sensor 11 on the display unit 4. Before performing yoga, the user U adjusts the relative position between the sensor 11 and the predetermined region so that the predetermined region (eg, yoga mat) is in the predetermined position in the image displayed on the display unit 4. May be. The detection apparatus 1 may display the mark (eg, guide, frame line, grid) indicating the predetermined position on the display unit 4 together with the image captured by the sensor 11. The user U performs yoga in the predetermined region in a state where the relative position between the sensor 11 and the predetermined region is adjusted.

  The storage unit 12 includes, for example, a volatile memory or a nonvolatile memory. The storage unit 12 stores the detection result of the sensor 11. For example, the storage unit 12 stores the detection result of the sensor 11 in association with the information of the timing at which the sensor 11 performs the detection. The timing information may be, for example, a number (for example, a frame number) assigned in the order of detection. Further, the timing information may be, for example, a time based on the timing at which the detection of the human body HM is executed (hereinafter referred to as a detection time). The detection time may be, for example, a time measured by a timing device (eg, an internal clock) built in the detection unit 2, or may be a time acquired from information received from the outside (eg, a standard radio wave indicating standard time). .

  For example, the detection unit 2 images the human body HM with the sensor 11 and causes the storage unit 12 to store data of the captured image (for example, pixel values such as RGB data and gray scale). The detection unit 2 can output the detection result of the sensor 11 to the outside. For example, the sensor 11 reads out the captured image data stored in the storage unit 12 and outputs the data to the outside. The detection unit 2 may output the detection result to the outside without going through the storage unit 12 (for example, in real time). In this case, the detection unit 2 may not include the storage unit 12.

  The processing device 3 is an information processing unit that processes information (for example, a detection result) output from the detection unit 2. The processing device 3 is communicably connected to the detection unit 2 by wire or wireless. The processing device 3 acquires (eg, receives) captured image data from the detection unit 2 as a detection result of the detection unit 2.

  Note that the processing device 3 may acquire the detection result or the processing result of the detection unit 2 without using communication with the detection unit 2. For example, the processing device 3 may acquire the detection result from the detection unit 2 via a storage medium such as a nonvolatile memory. For example, the storage unit 12 of the detection unit 2 may be a storage medium such as a memory card that is removable from the detection unit 2. For example, the processing device 3 may acquire the detection result of the detection unit 2 by being connected to the storage unit 12 removed from the processing device 3.

  The processing device 3 includes a processing unit 13, a comparison unit 14, an update unit 15, and a storage unit 16. The processing unit 13 processes information (eg, detection result, processing result) output from the detection unit 2. The processing unit 13 is, for example, an image processing unit that processes captured image data output from the detection unit 2. The processing unit 13 generates posture data FD representing the posture of the human body HM based on the detection result of the detection unit 2. For example, the processing unit 13 performs edge detection processing on the captured image of the detection unit 2, specifies the human body HM (eg, shape, contour) in the captured image, and generates posture data FD.

  In the present embodiment, the sensor 11 repeatedly detects the human body HM during a preset period (hereinafter referred to as a detection setting period). The processing unit 13 calculates the amount of change (eg, movement amount) of the human body HM between a plurality of captured images. For example, the processing unit 13 determines (recognizes) that the human body HM is stationary when the change amount of the human body HM is less than the threshold value between the previous captured image and the current captured image. For example, when the processing unit 13 determines that the human body HM is stationary, the processing unit 13 generates the posture data FD using the detection result of the sensor 11 that is the source of the determination.

  Note that the sensor 11 may capture one image, and the processing unit 13 may generate the posture data FD using this image. Further, the processing unit 13 may not determine whether the human body HM is stationary. In addition, the processing unit 13 may generate attitude data FD by averaging two or more images. For example, the processing unit 13 may generate the posture data FD based on the average (eg, addition average, weighted average) of the posture data obtained from the first image and the posture data obtained from the second image. .

  Further, the processing unit 13 may specify the part of the human body HM (eg, head, arm, torso, leg) in the captured image by pattern recognition and generate the posture data FD. Further, the processing unit 13 may detect the position of the human body HM in the captured image. For example, the processing unit 13 may represent the position of each part of the human body HM with two-dimensional coordinates. The two-dimensional coordinates may be, for example, data obtained by combining a pixel position in the horizontal scanning direction and a pixel position in the vertical scanning direction in the captured image. The processing unit 13 may generate part data in which the name and position of each part of the human body HM are associated with each other. The posture data FD may include, for example, table data obtained by collecting part data of a plurality of parts of the human body HM. The posture data FD may include information on the position of at least a part of the contour of the human body HM in the captured image. In addition, the posture data FD may include, for example, information on the position of a member (eg, an attachment, shoes, wear, a marker) attached to the human body HM. The processing unit 13 stores the generated posture data FD in the storage unit 16.

  The storage unit 16 includes, for example, a nonvolatile memory or a hard disk. The storage unit 16 stores initial data SD, teacher data TD, and posture data FD. The posture data FD stored in the storage unit 16 is information obtained by the processing unit 13 processing the detection result of the sensor 11 detecting the human body HM of the user U this time. In the present embodiment, the posture data FD is information representing a two-dimensional shape (eg, contour line) of the human body HM. The teacher data TD stored in the storage unit 16 is, for example, data selected from past posture data. The teacher data TD has the same data structure as the posture data FD. The teacher data TD may be data obtained by processing (eg, processing) the posture data FD. For example, the teacher data TD may be data obtained by extracting a part of the posture data FD.

  The initial data SD is, for example, a predetermined posture (e.g., crescent moon pose) taken by an object (e.g., instructor, robot) different from the user U or a user U's first set posture (e.g., initial posture). It is information to represent. The initial data SD includes, for example, the same data structure as the teacher data TD (eg, information representing at least a part of the contour of the object, position information of the part of the object).

  The past posture data does not exist when, for example, the user U taking a predetermined posture has not been detected once. The teacher data TD is not stored in the storage unit 12 in a state where the past posture data of the user U does not exist (for example, when the detection device 1 is shipped and before the detection device 1 is first used). In a state where the teacher data TD is not stored in the storage unit 12, the initial data SD is used for comparison with the posture data FD, for example, instead of the teacher data TD. The initial data SD may be used when the user U confirms a predetermined posture. For example, the processing device 3 may display the shape indicated by the initial data SD on the display unit 4.

  The processing device 3 may display the shape represented by the teacher data TD on the display unit 4 while the user U performs yoga. For example, the processing device 3 may cause the display unit 4 to display the shape represented by the teacher data TD based on a predetermined timing (eg, schedule). Further, the processing device 3 may cause the display unit 4 to display a detection result (eg, a captured image) of the sensor 11 that is the source of the teacher data TD. For example, the user U may perform yoga using the posture on the image as a model while viewing the image displayed on the display unit 4.

  The comparison unit 14 compares the posture data FD and the teacher data TD of the user acquired in the past. The comparison unit 14 reads the posture data FD and the teacher data TD from the storage unit 16, and compares the read posture data FD and the teacher data TD. For example, the comparison unit 14 compares the shape represented by the posture data FD (the current shape of the human body HM) with the shape represented by the teacher data TD (the shape of the past human body HM).

  For example, the comparison unit 14 compares the sizes of the shape represented by the posture data FD and the shape represented by the teacher data TD. For example, the comparison unit 14 appropriately corrects the posture data FD so that the difference in size between the shape represented by the posture data FD and the shape represented by the teacher data TD decreases. For example, the comparison unit 14 appropriately sets the posture data FD so that the shape (for example, the area surrounded by the outline of the human body HM) is the same between the shape represented by the posture data FD and the shape represented by the teacher data TD. Enlarge or reduce the shape to represent. The comparison unit 14 may compare the orientation between the shape represented by the posture data FD and the shape represented by the teacher data TD. For example, the comparison unit 14 may appropriately correct (eg, rotate) the direction of the shape represented by the posture data FD so that the shape represented by the posture data FD and the shape represented by the teacher data TD are the same. Good.

  Further, the comparison unit 14 calculates the similarity between the teacher data TD and the posture data FD. For example, the comparison unit 14 calculates a similarity (for example, a correlation coefficient) between the shape represented by the corrected posture data FD as described above and the shape represented by the teacher data TD. For example, the comparison unit 14 compares the entire shape represented by the teacher data TD with the entire shape represented by the posture data FD. The comparison unit 14 may compare the shape represented by the teacher data TD with the shape represented by the posture data FD for a part of the human body HM (eg, one part, a plurality of parts). The comparison unit 14 compares the shape represented by the teacher data TD and the shape represented by the posture data FD with respect to the relative positions (eg, the extension, bending, and angle between the upper arm and the forearm) of the human body HM. Good.

  The comparison unit 14 also generates comparison information indicating a comparison between the user U posture data FD and the user U teacher data TD. The comparison information includes, for example, an image (hereinafter referred to as a comparative image) indicating comparison between the posture data FD and the teacher data TD. The processing device 3 causes the display unit 4 to display the comparison image generated by the comparison unit 14. For example, the user U can input an update command indicating whether or not to update the teacher data TD to the input unit 5 with reference to the comparison image displayed on the display unit 4.

  The input unit 5 is used when the user U updates the teacher data TD based on the posture data FD. The input unit 5 receives an input of an update command indicating whether to update the teacher data TD using the posture data FD. The input unit 5 includes, for example, one or both of an operation device that is operated when an update command is input and a reception unit that receives the update command. The operation device includes, for example, at least one of a keyboard, a mouse, a trackball, a touch panel, and a voice input device (for example, a microphone). The receiving unit includes, for example, a wireless or wired communication device.

  Note that the input unit 5 may accept an input other than the update command. For example, the input unit 5 may receive an input of a command (for example, a detection start command) for starting detection by the detection unit 2. For example, the input unit 5 may include a voice input device, and the detection apparatus 1 may cause the detection unit 2 to start detection when a detection start command is input to the input unit 5 by voice. The input unit 5 is a contact-type input device such as a touch panel, and the detection apparatus 1 can detect the detection unit 1 after a delay time set in advance by a timer or the like when a detection start command is input to the input unit 5. The detection unit 2 may start detection.

  The update unit 15 updates the teacher data TD based on the update command input to the input unit 5 based on the comparison result of the comparison unit 14 and the posture data FD. Further, the update unit 15 registers the posture data FD as the teacher data TD when the teacher data TD does not exist. For example, when the user U taking a predetermined posture has never been detected, the teacher data TD does not exist. In such a case, the update unit 15 causes the storage unit 16 to store the posture data FD as teacher data TD based on, for example, a command from the user U or automatically. In the embodiment, the update of the teacher data TD includes an update from a state where the teacher data TD is not registered (eg, an initial state, a shipping state) to a state where the teacher data TD is registered.

  FIG. 2 is a diagram illustrating processing of the detection apparatus according to the present embodiment. In FIG. 2, reference symbol Im <b> 1 is a comparison image generated by the comparison unit 14. The comparison image Im1 includes, for example, information D1 (“Crescent Moon Pose” in the figure) indicating the name of the posture. The comparison image Im1 includes a partial image Im2 having a shape represented by the teacher data TD and a partial image Im3 having a shape represented by the posture data FD.

  In FIG. 2, the screen layout in the display unit 4 is a screen layout in which a display image (eg, Im3) based on the current posture data FD and a display image (eg, Im2) based on the teacher data TD are arranged. The screen layout can be changed, for example, when the user U changes the setting via the input unit 5. For example, the processing device 3 may display the display image (eg, Im3) based on the current posture data FD in the same size as the display image (eg, Im2) based on the teacher data TD. It may be displayed larger or smaller.

  In addition, one or both of a display image (eg, Im3) based on the current posture data FD and a display image (eg, Im2) based on the teacher data TD can be changed in the display position on the display unit 4. For example, the processing device 3 may display a display image (eg, Im2) based on the teacher data TD in the corner of the screen of the display unit 4 by the user U changing the setting via the input unit 5.

  For example, identification information D2 indicating that the partial image Im2 corresponds to the teacher data TD is attached. The identification information D2 is, for example, information on the detection time (“2017/02/10” in the figure) that is the basis of the teacher data TD. The identification information D2 may be other information, for example, character information such as “past posture”. For example, identification information D3 indicating that the partial image Im3 corresponds to the posture data FD is attached. The identification information D3 is, for example, detection time information (“2017/02/12” in the figure) that is the basis of the posture data FD. The identification information D3 may be other information, for example, character information such as “current posture”.

  In addition, the update unit 15 causes the display unit 4 to display a GUI (Graphical User Interface) button D4 and a pointer D5. The pointer D5 moves on the image displayed on the display unit 4 when the input unit 5 (see FIG. 1) is operated. The updating unit 15 monitors the input to the input unit 5, and when there is an input such as a click in a state where the pointer D5 is placed on the “Yes” of the button D4, an update command for updating the teacher data TD is issued. It is determined that it has been input. Further, the update unit 15 monitors the input to the input unit 5 and updates the teacher data TD not to be updated when there is an input such as a click in a state where the pointer D5 is arranged on the “No” of the button D4. It is determined that a command has been input.

  If the update unit 15 determines that an update command for updating the teacher data TD has been input, the update unit 15 updates and sets the teacher data TD stored in the storage unit 16. For example, the update unit 15 updates the teacher data TD by rewriting (overwriting) the teacher data TD stored in the storage unit 16 with the posture data FD.

  FIG. 3 is a diagram illustrating a modification of the processing of the comparison unit according to the present embodiment. 3A and 3B are examples of comparative images generated by the comparison unit 14, respectively. In FIG. 3A, the comparison image Im1 includes an image (eg, overlay image) in which the human body shape FG1 indicated by the teacher data TD and the human body shape FG2 indicated by the posture data FD are overlapped. The comparison unit 14 may represent the shapes FG1 and FG2 in different colors so that the shapes FG1 and FG2 can be distinguished. The comparison unit 14 may represent the shapes FG1 and FG2 with different brightness so that the shapes FG1 and FG2 can be distinguished. Further, the comparison unit 14 may represent the similarity D6 between the teacher data TD and the posture data FD on the comparison image Im1. The comparison unit 14 may represent the similarity D6 on the comparison image Im1 with character information such as a numerical value. Further, the comparison unit 14 may represent the degree of similarity by one or both of color (eg, red, blue) and brightness (eg, light, dark). The processing device 3 causes the user U to visually recognize the difference between the teacher data TD and the posture data FD by causing the display unit 4 to display an image in which one or both of color and brightness are changed in the shape FG1 and the shape FG2. May be notified in a recognizable manner.

  Further, the comparison unit 14 may represent the degree of similarity in color for the whole or a part of the shape FG2 represented by the posture data FD. For example, the comparison unit 14 may represent a portion with a relatively high similarity in blue and a portion with a relatively low similarity in red. Further, the comparison unit 14 may represent the degree of similarity with brightness for the whole or a part of the shape FG2 represented by the posture data FD. For example, the comparison unit 14 may represent a portion with a relatively high degree of similarity with a dark gradation and a portion with a relatively low degree of similarity with a bright gradation. Moreover, the comparison part 14 may change the presence or absence of a display with a site | part with a relatively high similarity, and a low site | part. For example, the comparison unit 14 may hide a portion with a relatively high similarity and selectively display a portion with a relatively low similarity.

  In FIG. 3B, the comparison image Im1 includes a shape FG2 indicated by the posture data FD and a graphic FG3 indicating a difference between the shape indicated by the teacher data TD in the shape FG2. For example, the comparison unit 14 extracts a portion different from the shape FG1 (see FIG. 3A) indicated by the teacher data TD from the shape FG2 indicated by the posture data FD. For example, the comparison unit 14 calculates a difference between the shape FG1 and the shape FG2 (eg, image calculation), and extracts a portion that is different between the shape FG1 and the shape FG2. The comparison unit 14 generates an image in which the figure FG3 representing the extracted portion (difference) is overlapped with the shape FG2 indicated by the posture data FD as the comparison image Im1. The processing device 3 displays such an image (for example, a superimposed image) on the display unit 4 to notify the user U of the difference between the teacher data TD and the posture data FD so as to be visually recognizable. Also good.

  Next, a detection method according to the embodiment will be described based on the operation of the detection apparatus 1 described above. FIG. 4 is a flowchart illustrating a detection method according to the embodiment. For each part of the detection apparatus 1, FIG. 1 is referred to as appropriate. In step S <b> 1, the detection unit 2 detects the human body HM of the user U. For example, the sensor 11 images the human body HM of the user U. For example, the sensor 11 repeatedly detects (eg, images) the human body HM during the detection setting period, and then ends the detection.

  The processing unit 13 generates posture data FD for at least a part of the human body HM based on the detection result of the sensor 11. The processing unit 13 may generate posture data during a period that overlaps the detection setting period. Further, the processing unit 13 may generate posture data after the detection setting period ends.

  In step S2, the comparison unit 14 compares the posture data FD of the human body with the teacher data TD. The comparison unit 14 may compare the posture data FD and the teacher data TD, and calculate the similarity (eg, correlation coefficient) between the shape represented by the posture data FD and the shape represented by the teacher data TD. The comparison unit 14 may compare the posture data FD and the teacher data TD, and calculate a difference between the shape represented by the teacher data TD and the shape represented by the posture data FD. The comparison unit 14 generates comparison information (for example, the comparison image Im1 in FIGS. 2 and 3) indicating comparison between the posture data FD and the teacher data TD. The processing device 3 causes the display unit 4 to display the comparison image Im1 generated by the comparison unit 14.

  In step S3, the update unit 15 determines whether to update the teacher data TD. When it is determined that the update command is input to the input unit 5, the update unit 15 determines to update the teacher data TD (step S3; Yes). If it is determined that the teacher data TD is to be updated (step S3; Yes), the update unit 15 updates the teacher data TD using the posture data FD in step S4. When the update unit 15 that does not update the teacher data TD determines (step S3; No), the detection apparatus 1 ends the series of processes.

  The processing unit 13 may be provided outside the processing device 3 (for example, the detection unit 2) in the detection device 1, or may be provided in an external device different from the detection device 1. For example, the detection device 1 may supply the detection result of the detection unit 2 to the external device, and acquire the posture data generated by the external device (eg, the processing unit 13). At least a part of the detection apparatus 1 may be a stationary apparatus such as a desktop personal computer. At least a part of the detection device 1 may be a portable information terminal such as a smartphone.

  The detection apparatus 1 according to the embodiment includes a detection unit 2 that detects the human body HM of the user U, posture data TD of the user U obtained from the detection result of the detection unit 2, and a teacher of the user U acquired in the past. A comparison unit 14 that compares the data TD, a processing unit 13 that generates user teacher data TD using the user attitude data FD based on the comparison result of the comparison unit 14, and an update unit 15. It does not have to be. The detection apparatus 1 may output at least part of posture data FD, teacher data TD, and comparison data (eg, comparison image data) between the posture data FD and the teacher data TD. Further, for example, the detection device 1 according to the present embodiment makes it possible to compare the posture of the user U with the posture of the user U in the past.

  The comparison unit 14 may not compare the posture data FD and the teacher data TD for a part of the human body HM of the user U. For example, the comparison unit 14 determines posture data (eg, position data of posture data) FD only for a comparison target portion set in advance based on the setting (comparison target portion setting information) stored in the storage unit 16. And teacher data (eg, part data of teacher data) TD may be compared. The above setting may be changeable by the designation of the user U or the like. For example, the above setting may be changeable so that a part where the target posture is almost achieved is excluded from the comparison target. Further, the above setting may be changeable so as to add a comparison target part.

[Second Embodiment]
A second embodiment will be described. In the present embodiment, the same components as those in the above-described embodiment are denoted by the same reference numerals, and the description thereof is omitted or simplified. FIG. 5 is a diagram illustrating the detection apparatus according to the present embodiment. In the present embodiment, the detection unit 2 optically detects the distance from each part of the human body HM. For example, the processing device 3 identifies each part of the human body HM based on the distance from each part of the human body HM to the detection unit 2, and uses the reference data in which the evaluation criteria for each part of the human body HM are determined in advance. Is calculated. The detection unit 2 may be, for example, a motion capture device or may detect a three-dimensional shape of the human body HM. The detection unit 2 includes, for example, a sensor 11, a processing unit 13, and a storage unit 12.

  The sensor 11 is a distance measuring device such as a depth sensor, for example. For example, the sensor 11 detects a distance from a predetermined viewpoint (eg, position of the sensor 11, viewpoint) to each point on the surface of the human body HM. The sensor 11 detects the distance by, for example, the TOF (time of flight) method. The sensor 11 outputs, for example, position information indicating a distance distribution as the detection result (eg, position information). The position information includes, for example, depth information (eg, depth image data, depth information). The sensor 11 outputs, for example, three-dimensional position information of a plurality of parts of the human body HM.

  The sensor 11 may be a device that detects a distance by a method other than the TOF method. The sensor 11 may be, for example, a device that includes a laser scanner (eg, laser range finder) and detects a distance by laser scanning. For example, the sensor 11 may be a device that projects a predetermined pattern onto an area through which the human body HM passes and measures the distance based on the detection result of the pattern. The sensor 11 may be, for example, a device that includes a phase difference sensor and detects a distance by a phase difference method. For example, the sensor 11 may be a device that detects a distance by a DFD (depth from defocus) method. The sensor 11 may be a device that detects a distance by triangulation using captured images obtained by capturing the human body HM from a plurality of viewpoints using, for example, a stereo camera. The sensor 11 may be a device that detects the human body HM by combining two or more of the plurality of detection methods described above. The sensor 11 may include a sensor that detects a distance and the image sensor described in the first embodiment.

  FIG. 6 is a diagram illustrating processing of the processing unit according to the embodiment. For example, the sensor 11 (see FIG. 5) outputs data of the depth image Im5. In the depth image Im5, the pixel value of each pixel represents, for example, the distance between the sensor 11 and a point in the real space corresponding to the position of the pixel on the depth image Im5. For example, the depth image Im5 in FIG. 6 is represented in gray scale, and the dark portion is more distant from the sensor 11 than the bright portion.

  The processing unit 13 (see FIG. 5) detects (calculates) the position of the human body HM based on the detection result of the sensor 11. For example, the processing unit 13 generates (calculates) shape information representing the shape of the human body HM by processing the data of the depth image Im5 (eg, perspective conversion processing, projective conversion processing). The shape information includes, for example, point cloud data. The point cloud data includes, for example, a plurality of point data on the surface of the human body HM. For example, the plurality of point data are each represented by three-dimensional coordinates.

  The processing unit 13 may calculate surface information as shape information. The surface information is, for example, polygon data, vector data, draw data, or the like. The surface information includes, for example, coordinates of a plurality of points and connection information between the plurality of points. The connection information includes, for example, information that associates points at both ends of a line corresponding to a ridge line (eg, edge) of an object (eg, human body HM). The connection information includes, for example, information that associates a plurality of lines corresponding to the contour of the surface on the object.

  For example, the processing unit 13 estimates a surface between a point selected from a plurality of points included in the point cloud data and its neighboring points, and converts the point cloud data into polygon data having plane information between the points. May be. For example, the processing unit 13 converts the point cloud data into polygon data by an algorithm using a least square method. This algorithm may be, for example, an algorithm published in the point cloud processing library.

  Based on the detection result of the sensor 11, the processing unit 13 generates position information PD of a characteristic part (eg, characteristic point) of the human body HM. The characteristic part of the human body HM is, for example, a part that can be distinguished from other parts of the human body HM. The characteristic part of the human body HM includes, for example, at least one of an end portion of the human body HM, a joint, or an intermediate portion between the end portion and the joint or between the two joints.

  For example, the processing unit 13 generates the position information PD of the characteristic part by a recognition process (eg, pattern recognition, shape recognition, skeleton recognition) using point cloud data. The position information PD of the characteristic part includes, for example, coordinates of a point representing the characteristic part (eg, three-dimensional coordinates). For example, the processing unit 13 calculates the coordinates of the point representing the characteristic part by the above recognition processing. For example, the processing unit 13 generates data that associates the name of the characteristic part and the position of the characteristic part for each of the characteristic parts. For example, the processing unit 13 generates position information PD (for example, table data) including position information of each of one or more preset characteristic parts. The position information PD includes, for example, position information of a plurality of characteristic parts obtained from detection results at the same detection timing.

  Returning to the description of FIG. 5, the detection unit 2 outputs the position information generated by the processing unit 13 to the outside. For example, the processing unit 13 stores the generated position information in the storage unit 12. For example, the detection unit 2 outputs the position information stored in the storage unit 12 to the processing device 3. The processing device 3 stores the position information output from the detection unit 2 in the storage unit 16. In the present embodiment, the storage unit 16 stores reference data RD, teacher data TD, and posture data FD. The storage unit 16 may store the initial data SD shown in FIG. 1 or may not store the initial data SD.

  The reference data RD is information (definition data, reference data) that defines a target posture (eg, yoga crescent pose). The reference data RD is information that predetermines evaluation criteria for each part of the human body HM. The processing device 3 can register the target posture, for example, according to a command from the user U. For example, the user U operates the input unit 5 to cause the processing device 3 to acquire information related to the target posture. The processing device 3 acquires (eg, downloads) data relating to the target posture (reference data RD, initial data SD in FIG. 1) via, for example, the Internet line.

  Note that the processing device 3 may acquire information on the target posture from a storage medium such as an optical disk (eg, DVD-ROM) or a USB memory. Further, the processing device 3 may automatically acquire information on the target posture without receiving a command from the user U based on a setting such as automatic update. Further, the processing device 3 does not have to acquire information about a new target posture by executing processing regarding the target posture stored in the storage unit 16 in advance.

  The comparison unit 14 evaluates the posture represented by the posture data FD of the user U using the reference data RD. The comparison unit 14 calculates the similarity between the teacher data TD and the posture data FD using the reference data RD. FIG. 7 is a diagram illustrating processing of the comparison unit according to the present embodiment. FIG. 7A conceptually shows the reference data RD. The reference data RD is information in which a part name of the human body HM, an evaluation item, and a weighting coefficient (weighting coefficient) are associated with each other. For example, for the head part name, the evaluation item is the relative position of the head and neck, and the weighting factor is C1. For the part name of the elbow, the evaluation item is the relative position between the shoulder-elbow (upper arm) and the elbow-wrist (forearm), and the weighting factor is Ci. The comparison unit 14 evaluates the posture for each preset part.

  FIG. 7B conceptually shows an evaluation process for evaluating the posture for each part. When the posture of the elbow is evaluated, the comparison unit 14 uses the posture data FDa, and the relative position (eg, angle θ1) between the upper arm Qc1 from the shoulder Qa1 to the elbow Qb1 and the forearm Qe1 from the elbow Qb1 to the wrist Qd1. Is calculated. Further, the comparison unit 14 calculates the relative position (eg, angle θ2) between the upper arm Qc2 from the shoulder Qa2 to the elbow Qb2 and the forearm Qe2 from the elbow Qb2 to the wrist Qd2 using the teacher data TD. The comparison unit 14 compares the angle θ1 and the angle θ2 and calculates an elbow evaluation value. For example, the comparison unit 14 calculates a value obtained by multiplying the difference between the angle θ1 and the angle θ2 by the weight coefficient Ci as the evaluation value.

  The comparison unit 14 evaluates the posture by distinguishing the direction in which the posture data FD deviates from the teacher data TD. For example, when the posture is desired to extend the elbow, the comparison unit 14 relatively increases the evaluation value when the angle θ1 is closer to 180 ° than the angle θ2. For example, the comparison unit 14 calculates an evaluation value by multiplying a value obtained by subtracting the angle θ2 from the angle θ1 by a weighting factor Ci. In FIG. 7B, the angle θ1 is closer to 180 ° than the angle θ2, (θ1−θ2) is positive, and the evaluation value is a positive value. When the elbow Qb3 is bent more than the teacher data TD as in the posture data FDb, the angle θ3 between the upper arm Qc3 from the shoulder Qa3 to the elbow Qb3 and the forearm Qe3 from the elbow Qb3 to the wrist Qd3 is greater than the angle θ2. Becomes smaller. In this case, (θ3−θ2) is negative and the evaluation value is negative.

  In the reference data RD, an arithmetic expression of a value calculated from the teacher data TD and a value calculated from the posture data FD is predetermined for each part. The comparison unit 14 calculates an evaluation value by distinguishing the direction in which the posture data FD deviates from the teacher data TD by performing a calculation based on an arithmetic expression defined in the reference data RD. The comparison unit 14 calculates an evaluation value for each preset part, and calculates the similarity by integrating the evaluation values at the preset part.

  The position of the first part (eg, shoulder Qa1) in the posture data FD may be shifted from the position of the first part (eg, shoulder Qa2) in the teacher data TD. In this case, the position of the second part (eg, elbow, wrist) on the terminal side of the human body HM with respect to the first part is shifted between the teacher data TD and the posture data FD due to the displacement of the first part. As described above, when the teacher data TD and the posture data FD are compared using the relative positions of the parts of the human body, it is possible to reduce the influence of the positional deviation.

  Note that the evaluation item and the weighting coefficient may not be set for a part of the human body based on the target posture. For example, in a yoga pose, an evaluation item and a weighting coefficient may not be set for a portion that may be stretched or bent. Further, the reference data RD may not include a weighting coefficient. For example, the comparison unit 14 may calculate the similarity by integrating or averaging the values calculated for each evaluation item.

  Note that the comparison unit 14 may compare the posture data FD and the teacher data TD without using the reference data RD. The comparison unit 14 does not have to calculate the similarity between the posture data FD and the teacher data TD. As described in the first embodiment, the comparison unit 14 may generate a comparison image indicating a comparison between the posture data FD and the teacher data TD. The processing unit 13 may be provided outside the detection unit 2 (for example, the processing device 3) in the detection device 1 or may be provided in an external device different from the detection device 1.

[Third Embodiment]
A third embodiment will be described. In the present embodiment, the same components as those in the above-described embodiment are denoted by the same reference numerals, and the description thereof is omitted or simplified. FIG. 8 is a diagram illustrating the detection device according to the embodiment. In the present embodiment, the detection unit 2 repeatedly detects the human body HM. The comparison unit 14 compares each of the plurality of posture data (FD1 to FDj) obtained from the detection result of the detection unit 2 repeatedly detecting the human body HM (see FIG. 1) with the teacher data TD. The update unit 15 updates the teacher data TD to posture data selected from a plurality of posture data (FD1 to FDj) based on the comparison result of the comparison unit 14. For example, the comparison unit 14 generates a comparison image of the best posture data and teacher data of the user U (see FIG. 1) in a predetermined period (eg, a period of performing a series of yoga, one day, one hour). For example, the user U can specify whether or not to update the teacher data with the posture data based on the comparison between the best posture data and the teacher data.

  The detection unit 2 attaches identification information (for example, ID) to the detection result obtained by each detection, and supplies it to the processing unit 13. This identification information is, for example, a number (eg, frame number) assigned in the order in which detection results are obtained. For example, in FIG. 8, the detection result output by the detection unit 2 has an ID of j (j is an integer of 1 or more).

  The processing unit 13 processes the detection result in the detection unit 2 (for example, the sensor 11) to generate posture data. The processing unit 13 attaches identification information (for example, ID) to the generated posture data and outputs it. This identification information is, for example, an identification number assigned in the order in which posture data is obtained. For example, in FIG. 8, the posture data output from the processing unit 13 has identification information j (j is an integer of 1 or more).

  The storage unit 16 stores the posture data output by the processing unit 13 in association with the identification information. For example, in FIG. 8, the posture data FD includes a plurality of posture data whose identification information is 1 to j. In the following description, when a plurality of posture data are distinguished, a suffix (1 to j) corresponding to the identification information is given. For example, the posture data FDj is posture data whose identification information is j.

  The comparison unit 14 reads each of the plurality of posture data FD1 to FDj from the storage unit 12, and compares it with the teacher data TD using the reference data RD. The comparison unit 14 calculates the similarity for each of the plurality of posture data FD1 to FDj using the reference data RD. For example, the comparison unit 14 reads the posture data FDj, the teacher data TD, and the reference data RD from the storage unit 12, and calculates the similarity between the posture data FDj and the teacher data TD using these data. The comparison unit 14 stores the calculated similarity in the storage unit 16 in association with the identification information of the posture data FDj. For example, the storage unit 16 stores similarity data CD including the similarity associated with the identification information. The similarity data CD is, for example, table data in which data obtained by combining identification information and similarity is collected.

  In the present embodiment, the detection device 1 includes a selection unit 21. The selection unit 21 selects posture data having a relatively high similarity with the teacher data TD from the plurality of posture data FD1 to FDj. For example, the processing unit 13 sends information (eg, signal, flag) indicating the end of the similarity calculation process to the selection unit 21 after the similarity calculation processing for calculating the similarity for a plurality of scheduled posture data is completed. Supply. When the selection unit 21 receives information indicating the end of the similarity calculation process from the comparison unit 14, the selection unit 21 reads the similarity data CD from the storage unit 12. The selection unit 21 selects (specifies) identification information of posture data having a relatively high similarity in the similarity data CD.

  Here, it is assumed that the similarity of the posture data FDj is the highest in the similarity data CD. The selection unit 21 supplies the selected identification information (ID = j) to the comparison unit 14. The comparison unit 14 reads the posture data FDj corresponding to the identification information (ID = j) supplied from the selection unit 21 from the storage unit 16. The comparison unit 14 generates a comparison image Im1 (see FIGS. 2 and 3) that shows a comparison between the posture data FDj and the teacher data TD. The display unit 4 displays an image based on the posture data FD selected by the selection unit 21. For example, the display unit 4 displays the comparison image Im1 (see FIGS. 2 and 3) generated by the comparison unit 14 based on the posture data FD selected by the selection unit 21.

  Note that the selection unit 21 may select two or more posture data having a relatively high similarity to the teacher data TD from the plurality of posture data FD1 to FDj. For example, the selection unit 21 may select identification information of a predetermined number of posture data from the similarity data CD in descending order of similarity of the posture data. The selection unit 21 may extract the identification information of the posture data that satisfies a predetermined condition (for example, a condition that the similarity is equal to or greater than the threshold) by comparing the similarity and the threshold.

  Note that the comparison unit 14 may generate a comparison image indicating comparison between two or more posture data (eg, posture data FD1 and FD2). For example, the comparison unit 14 may generate a comparison image in which shapes represented by two or more posture data are arranged. Further, the comparison unit 14 may generate a comparison image indicating comparison with two or more posture data (eg, posture data FD1, FD2) and the teacher data TD. For example, the comparison unit 14 may generate a comparative image in which the shape represented by two or more posture data and the shape indicated by the teacher data TD are arranged.

  Further, the comparison unit 14 may generate a comparison image indicating a comparison with the teacher data TD for each of the two or more posture data selected by the selection unit 21. The processing device 3 may display a plurality of comparison images generated by the comparison unit 14 side by side on the display unit 4. Further, the processing device 3 may cause the display unit 4 to sequentially display a plurality of comparison images generated by the comparison unit 14.

[Fourth Embodiment]
A fourth embodiment will be described. In the present embodiment, the same components as those in the above-described embodiment are denoted by the same reference numerals, and the description thereof is omitted or simplified. FIG. 9 is a diagram illustrating a detection device according to the present embodiment. FIG. 10 is a diagram illustrating processing of the comparison unit according to the present embodiment. In the present embodiment, the detection apparatus 1 compares the posture data FD with a plurality of teacher data (TD1 to TDk). For example, the detection apparatus 1 compares the posture of the user U with the teacher data for the posture of the cat and the teacher data for the posture of the cow.

  In the present embodiment, the teacher data TD includes a plurality of teacher data. Each of the plurality of teacher data is stored in the storage unit 16 in association with the identification information. In the following description, when a plurality of teacher data is distinguished, a suffix (1 to k) corresponding to the identification information is added. For example, the teacher data TDk is posture data of the user U whose identification information is k.

  The plurality of teacher data TD1 to TDk includes teacher data TD1 for the first posture F1 (see FIG. 10A) and teacher data TD2 for the second posture F2 (see FIG. 10B). The first posture F1 (eg, first target posture, cat pose) is different from the second posture F2 (eg, second target posture, cow pose). The order in which the user takes the first posture F1 and the order in which the user takes the second posture F2 are determined in advance by the user's designation, for example. For example, it is predetermined that the first posture and the second posture are taken continuously. The storage unit 16 stores in advance timing data (schedule information) that defines the order and time in which the user takes each posture.

  The detection unit 2 repeatedly detects the user's human body over a period in which the user takes the first posture F1 and a period in which the user takes the second posture F2. For example, the detection unit 2 repeatedly detects the user's human body during a period determined based on the timing data stored in the storage unit 16. The detection unit 2 attaches identification information (for example, ID) to the detection result obtained by each detection, and supplies it to the processing unit 13. The processing unit 13 processes the detection result from the detection unit 2 and generates posture data. The processing unit 13 attaches identification information (for example, ID) to the generated posture data and outputs it.

  The storage unit 16 stores the posture data output by the processing unit 13 in association with the identification information. The posture data FD1 corresponds to, for example, the first posture F1 (eg, cat pose) in FIG. The posture data FDj corresponds to, for example, the second posture F2 (eg, cow pose) in FIG. The posture data FD2 is posture data corresponding to the posture in the process of moving from the first posture F1 to the second posture F2, for example.

  The comparison unit 14 reads the plurality of posture data FD1 to FDj from the storage unit 12, respectively. The comparison unit 14 performs comparison using different teacher data for each user posture data. The comparison unit 14 compares each of the plurality of posture data FD1 to FDj with the teacher data TD1 for the first posture F1 and the teacher data TD2 for the second posture F2. For example, as illustrated in FIG. 10B, the comparison unit 14 compares the posture data FD1 and the teacher data TD1. Further, the comparison unit 14 compares the posture data FD1 and the teacher data TD2. For example, as illustrated in FIG. 10C, the comparison unit 14 compares the posture data FDj and the teacher data TD1. Further, the comparison unit 14 compares the posture data FDj and the teacher data TD2.

  For example, the comparison unit 14 calculates the similarity between each of the plurality of posture data FD1 to FDj and the teacher data TD1. The comparison unit 14 generates a comparative image using posture data having a relatively high similarity to the teacher data TD1 among the plurality of posture data FD1 to FDj. For example, the posture data FD1 has a higher degree of similarity with the teacher data TD1 than the posture data FDj. In this case, the comparison unit 14 generates a comparison image indicating a comparison between the posture data FD1 and the teacher data TD1.

  For example, the comparison unit 14 calculates the similarity between each of the plurality of posture data FD1 to FDj and the teacher data TD2. The comparison unit 14 generates a comparative image using posture data having a relatively high similarity to the teacher data TD2 among the plurality of posture data FD1 to FDj. For example, the posture data FDj has a higher degree of similarity with the teacher data TD2 than the posture data FD1. In this case, the comparison unit 14 generates a comparison image indicating the comparison between the posture data FDj and the teacher data TD2.

[Fifth Embodiment]
A fifth embodiment will be described. In the present embodiment, the same components as those in the above-described embodiment are denoted by the same reference numerals, and the description thereof is omitted or simplified. FIG. 11 is a diagram illustrating a detection device according to the present embodiment. In the present embodiment, the detection device 1 notifies the user U of the timing of posture change based on the timing of breathing of the human body HM. The detection device 1 includes a timing detection unit 22, a schedule setting unit 23, and a notification unit 24.

  The timing detection unit 22 detects the breathing timing of the human body HM based on the detection result of the detection unit 2. For example, the detection unit 2 repeatedly detects the user U performing breathing before yoga or in the initial posture of yoga. The processing unit 13 generates posture data of the user U performing breathing based on the detection result of the detection unit 2. The processing unit 13 stores the generated posture data in the storage unit 16. As shown in FIGS. 8 and 9, posture data FD (for example, a plurality of posture data FD1 to FDj in FIG. 8) obtained by the detection unit 2 repeatedly detecting the human body HM is stored in the storage unit 16. The

  The timing detection unit 22 reads out the posture data FD of the user U performing breathing from the storage unit 16. For example, the timing detection unit 22 calculates a time change of the posture data FD. For example, the timing detection unit 22 detects the movement of a part (eg, chest) of the human body HM accompanying the breathing of the user U. The timing detection unit 22 estimates the respiration cycle of the user U based on, for example, the movement cycle of the part (eg, chest, abdomen) of the human body HM accompanying the respiration of the user U. The timing detection unit 22 stores the estimated respiration cycle in the storage unit 16.

  The schedule setting unit 23 generates timing data LD for posture change of the human body HM based on the breathing cycle estimated by the timing detection unit 22. For example, the schedule setting unit 23 determines, for example, the time for taking the first posture F1 shown in FIG. 10 based on the breathing cycle. The time for taking the first posture F1 is, for example, the time for maintaining one pose, the time for which it is recommended to take the first posture F1, and the like. For example, the schedule setting unit 23 determines the time taken for the first posture F1 to be an integral multiple of the respiratory cycle.

  The schedule setting unit 23 generates the timing data LD based on, for example, the determined time of each posture and the planned posture order. The timing data LD is, for example, information representing the timing of taking each posture for each of a plurality of scheduled yoga postures. For example, the timing data LD includes information representing the start timing and end timing of the period in which the first posture F1 is taken as the time from the start of yoga. The schedule setting unit 23 stores the generated timing data LD in the storage unit 16.

  The timing detection unit 22 stores the breathing cycle in the storage unit 16 at least once. In the subsequent processing, when the timing detection unit 22 does not detect the respiration cycle, the schedule setting unit 23 reads the respiration cycle stored in the storage unit 16 and generates the timing data LD.

  The notification unit 24 notifies the timing of the posture change of the human body HM based on the detection result of the timing detection unit 22. For example, the notification unit 24 notifies the timing of the posture change of the human body HM based on the timing data LD generated by the schedule setting unit 23 based on the detection result of the timing detection unit 22. For example, the processing device 3 supplies a notification signal to the notification unit 24 based on the timing data LD. For example, the notification unit 24 notifies the timing of the posture change of the human body by voice. The notification unit 24 includes, for example, an audio reproduction device, an amplifier, a speaker, and the like. The notification signal is voice data representing sounds such as a voice guide such as “Please take the next posture”, music, and chimes. The notification unit 24 performs notification by reproducing the sound represented by the notification signal supplied from the processing device 3.

  Note that the detection apparatus 1 may not include the timing detection unit 22. For example, the user U may input his / her breathing timing to the processing device 3 through the input unit 5. The schedule setting unit 23 may generate timing data LD based on the breathing timing input from the user U. Further, the detection device 1 may not include the schedule setting unit 23. For example, the user U may set timing for operating the input unit 5 to perform a plurality of postures.

  Note that the notification unit 24 may notify the timing of posture change based on the timing data LD set and edited by the user U, for example. Moreover, the alerting | reporting part 24 may alert | report other than an audio | voice. For example, the alerting | reporting part 24 may alert | report by light (light lighting), a vibration, etc. Moreover, the display part 4 may serve as the alerting | reporting part 24, and may alert | report by displaying an image. Further, the detection device 1 may not include the notification unit 24. Further, the detection system SYS may include a notification unit 24 outside the detection device 1.

[Sixth Embodiment]
A sixth embodiment will be described. In the present embodiment, the same components as those in the above-described embodiment are denoted by the same reference numerals, and the description thereof is omitted or simplified. FIG. 12 is a diagram illustrating a detection device according to the present embodiment. For example, the detection device 1 can automatically update the teacher data according to the determination of the determination unit 25 without receiving the update instruction of the user U.

  In the present embodiment, the detection device 1 includes a determination unit 25. The determination unit 25 determines whether to update the teacher data TD based on the comparison result of the comparison unit 14 (executes determination processing). For example, the comparison unit 14 performs the determination process based on the similarity between the posture data FD and the teacher data TD. For example, the determination unit 25 compares the similarity between the posture data FD and the teacher data TD and a threshold value. The determination unit 25 determines to update the teacher data TD when the similarity is equal to or greater than the threshold. When the determination unit 25 determines to update the teacher data TD, the determination unit 25 supplies an update command (for example, a setting signal) to the update unit 15. When receiving an update command (setting command) from the determination unit 25, the update unit 15 updates the teacher data TD (for example, automatically updates) using the posture data FD.

[Seventh Embodiment]
A seventh embodiment will be described. In the present embodiment, the same components as those in the above-described embodiment are denoted by the same reference numerals, and the description thereof is omitted or simplified. FIG. 13 is a diagram illustrating a detection device according to the present embodiment. The detection device 1 outputs, for example, the teacher data TD and the posture data FD to an external device (eg, the processing device 30), and updates the teacher data TD based on an update command supplied from the external device (eg, the processing device 30). To do.

  In the present embodiment, the detection device 1 includes a communication unit 26. The communication unit 26 transmits at least a part of the information (eg, teacher data TD, posture data FD, comparison image data) stored in the storage unit 16 to the outside of the detection device 1 (eg, the processing device 30). The communication unit 26 is connected to the processing device 30 via, for example, an Internet line.

  The processing device 30 is a cloud computer, for example. The processing device 30 may determine the quality of the posture. The processing device 30 may determine whether to update the teacher data TD based on the posture data FD received from the detection device 1 (for example, the communication unit 26). The processing device 30 may transmit an update command to the detection device 1 (for example, the communication unit 26) based on the determination result of whether or not to update the teacher data TD. The update unit 15 may update the teacher data TD based on the update command from the processing device 30.

  The processing device 30 may be a computer that can be operated by an operator such as an instructor. For example, the processing device 30 may receive the data of the comparison image from the detection device 1 and display the comparison image on the display unit. The instructor may determine whether or not to update the teacher data TD or whether or not to update the teacher data TD based on the comparison image displayed on the display unit. The processing device 30 may transmit an update command to the detection device 1 according to an instructor's command. The update unit 15 may update the teacher data TD based on the update command from the processing device 30.

  As described above, the update unit 15 may update the teacher data TD by receiving an update command from a device external to the detection device 1. In this case, the detection apparatus 1 may not include the determination unit 25 illustrated in FIG. Moreover, the detection apparatus 1 may be connectable to a social network (SNS) etc. via the communication part 26. FIG. The detection apparatus 1 may post and publish at least a part of the information stored in the storage unit 16 on the social network based on the instruction of the user U. Further, the detection apparatus 1 may receive one or both of the initial data SD of FIG. 1 and the reference data of FIG. 5 via the communication unit 26. The detection apparatus 1 may acquire, for example, information on a target posture that is not stored in the storage unit 16 (eg, initial data SD and reference data RD on a new posture) via the communication unit 26.

[Eighth Embodiment]
An eighth embodiment will be described. In the present embodiment, the same components as those in the above-described embodiment are denoted by the same reference numerals, and the description thereof is omitted or simplified. FIG. 14 is a diagram illustrating a detection device according to the present embodiment. In the present embodiment, the detection apparatus 1 specifies posture data used as teacher data. Instead of overwriting the teacher data with the posture data, the detection apparatus 1 updates the teacher data by changing the designation of the posture data used as the teacher data.

  In the present embodiment, the storage unit 16 stores user U posture data FD and setting data VD for specifying posture data FD used as user U teacher data. Posture data FD includes a plurality of posture data FD1 to FDn. Each of the plurality of posture data FD1 to FDn is provided with identification information (eg, ID). Here, it is assumed that the posture data FDn whose identification information is n is the latest posture data. In the present embodiment, the detection apparatus 1 uses posture data selected from a plurality of posture data FD1 to FDn-1 acquired before the posture data FDn as teacher data.

  The setting data VD includes setting values for specifying posture data used as teacher data among the plurality of posture data FD1 to FDn. For example, in FIG. 14, the posture data identification information (ID) used for the teacher data is defined as “m”. The comparison unit 14 reads the setting data stored in the storage unit 16 and acquires identification information (ID = m) that specifies teacher data. The comparison unit 14 reads out the posture data FDm used as teacher data and the posture data FDn to be compared from the storage unit 16. The comparison unit 14 compares the posture data FDn and the posture data FDm (teacher data).

  Further, the comparison unit 14 supplies the identification information (ID = n) of the posture data FD to be compared to the update unit 15. When receiving an update command from the input unit 5, the update unit 15 updates the setting data using the identification information supplied from the comparison unit 14. The update unit 15 updates the teacher data from the posture data FDm to the posture data FDn, for example, by rewriting the device-specific information specifying the teacher data from m to n. As described above, when updating the teacher data, the update unit 15 replaces the teacher data TD stored in the storage unit 16 with the posture data FD, and specifies information (e.g., setting data VD, identification data). Information) may be updated.

[Ninth Embodiment]
A ninth embodiment will be described. In the present embodiment, the same components as those in the above-described embodiment are denoted by the same reference numerals, and the description thereof is omitted or simplified. FIG. 15 is a diagram illustrating a detection system according to the embodiment. The detection system SYS includes a detection device 1 and a control device 32. In the present embodiment, the detection device 1 includes a processing device 3 and a plurality of detection units 2. The plurality of detection units 2 are arranged so as to surround the user U so that the user U can be detected from different directions (for example, different directions with respect to the user U). The processing device 3 processes the detection results output from each of the plurality of detection units 2.

  The control device 32 controls each part (for example, the detection unit 2 and the processing device 3) of the detection system SYS. The control device 32 supplies a control signal to each of the plurality of detection units 2. The control device 32 causes the detection unit 2 to perform detection based on the control signal. For example, the control device 32 controls the timing at which each of the plurality of detection units 2 executes detection. Further, the control device 32 supplies a control signal to the processing device 3. The control device 32 causes the processing device 3 to execute processing according to the control signal. For example, the control device 32 controls the timing at which the processing device 3 executes processing with respect to the timing detected by the detection unit 2. For example, the control device 32 causes the processing device 3 to process the detection result of the detection unit 2 after the detection unit 2 finishes the scheduled detection. The control device 32 causes the processing device 3 to output at least a part of the processing results (eg, comparison image data, posture data FD, teacher data TD in FIG. 1).

  In addition, the number of the detection parts 2 with which the detection apparatus 1 is provided may be one, and may be two or more. Moreover, the number of the detection apparatuses 1 with which the detection system SYS is provided may be one, or two or more. The control device 32 may be provided in the same device (for example, a computer) as the processing device 3. Further, the detection system SYS may not include the control device 32. The detection system SYS may be configured by only the detection device 1 or may include a device separate from the detection device 1 and the control device 32.

  In the above-described embodiment, the processing device 3 includes a computer system, for example. The processing device 3 reads the detection program stored in the storage unit 16 and executes various processes according to the detection program. This detection program is based on, for example, comparing the posture data of the human body obtained from the detection result of detecting the human body of the user with the teacher data of the user acquired in the past, and the posture data and the comparison. And updating the teacher data using the posture data. This detection program may be provided by being recorded on a computer-readable storage medium. The detection program may include application software such as a game. For example, the above detection program may be stored in a server or the like and provided by being downloaded to a computer system (eg, a smartphone). The processing device 3 may be provided separately from at least a part of the detection device 1 (for example, the detection unit 2).

  The technical scope of the present invention is not limited to the aspects described in the above-described embodiments. One or more of the requirements described in the above embodiments and the like may be omitted. In addition, the requirements described in the above-described embodiments and the like can be combined as appropriate. In addition, as long as it is permitted by law, the disclosure of all documents cited in the above-described embodiments and the like is incorporated as a part of the description of the text.

DESCRIPTION OF SYMBOLS 1 ... Detection apparatus, 2 ... Detection part, 3 ... Processing apparatus, 4 ... Display part, 5 ... Input part, 13 ... Processing part, 14 ... Comparison part, 15 ... update unit, 16 ... storage unit, 21 ... selection unit, 22 ... timing detection unit, 24 ... notification unit, 25 ... determination unit,
31 ... Detection system, FD ... Posture data, HM ... Human body, TD ... Teacher data

Claims (23)

A detection unit for detecting a user's human body;
Based on the comparison result obtained by comparing the user posture data obtained from the detection result of the detection unit and the user teacher data acquired in the past, the teacher data is obtained using the user posture data. An update unit for updating.   The detection device according to claim 1, further comprising: a comparison unit that compares the posture data of the human body obtained from the detection result of the detection unit with the teacher data of the user acquired in the past and outputs the comparison result. . The comparison unit calculates a similarity between the teacher data and the posture data;
The detection device according to claim 2, wherein the update unit updates the teacher data based on the similarity.   The detection device according to claim 2, wherein the comparison unit calculates the similarity using reference data in which an evaluation reference for each part of the human body is determined in advance.   5. The detection device according to claim 2, wherein the comparison unit compares the posture data with each of the teacher data with respect to the first posture and the teacher data with respect to the second posture. 6. .   The detection device according to any one of claims 2 to 5, wherein the comparison unit performs comparison using the teacher data that is different for each posture data of the user. The comparison unit compares each of the plurality of posture data obtained from the detection result of the detection unit repeatedly detecting the human body with the teacher data,
The detection according to any one of claims 2 to 6, wherein the update unit updates the teacher data to posture data selected from the plurality of posture data based on a comparison result of the comparison unit. apparatus.   The detection device according to claim 7, further comprising a selection unit that selects posture data having a relatively high similarity to the teacher data from the plurality of posture data.   The detection device according to claim 8, further comprising a display unit that displays an image based on the posture data selected by the selection unit.   The detection device according to claim 2, further comprising a display unit that displays an image based on a comparison result of the comparison unit.   The detection device according to any one of claims 1 to 10, further comprising an input unit for the user to update the teacher data based on the posture data.   The detection device according to any one of claims 1 to 11, further comprising a determination unit that determines whether or not to update the teacher data based on the comparison result. A storage unit for storing the teacher data;
The detection device according to claim 1, wherein the update unit rewrites the teacher data stored in the storage unit with the posture data.   The detection device according to claim 1, further comprising a processing unit that generates the posture data based on a detection result of the detection unit.   The detection device according to claim 1, further comprising a timing detection unit that detects a timing of breathing of the human body based on a detection result of the detection unit.   The detection device according to claim 15, further comprising a notification unit that notifies the timing of the posture change of the human body based on a detection result of the timing detection unit.   The detection device according to any one of claims 1 to 16, further comprising a notification unit that notifies the timing of the posture change of the human body by voice.   The detection device according to any one of claims 1 to 17, wherein the posture data includes posture data related to yoga of the user. A detection unit for detecting a user's human body;
A comparison unit that compares the posture data of the user obtained from the detection result of the detection unit and the teacher data of the user acquired in the past;
And a processing unit that generates teacher data of the user using the posture data of the user based on a comparison result of the comparison unit.   A detection system comprising the detection device according to any one of claims 1 to 17.   An update unit that updates the teacher data using the posture data based on a comparison result between the posture data of the human body obtained from the detection result of detecting the human body of the user and the teacher data of the user acquired in the past A processing apparatus comprising: Detecting the user's body,
And updating the teacher data using the posture data based on a comparison result between the posture data of the human body obtained from the detection and the teacher data of the user acquired in the past. On the computer,
Updating the teacher data using the posture data based on a comparison result between the posture data of the human body obtained from the detection result of detecting the human body of the user and the teacher data of the user acquired in the past. Detection program to be executed.