Moving body, control device, and control method
By combining posture driving and guiding sound on the mobile body, users are guided to move their faces in the face registration process, solving the problem of unintuitive face registration in the prior art, and improving the intuitiveness of operation and the accuracy of face recognition.
Patent Information
- Application Number
- CN202380029784.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-04-01
- Filing Date
- 2023-03-15
- Publication Date
- 2024-11-08
AI Technical Summary
In the prior art, users need to guide characters and images on the display during face registration processing, resulting in the face registration process being not intuitive enough and the user has difficulty in operating.
The posture control part and the guidance sound control part are adopted to drive and guide the sound through the posture of the moving body to guide the user to move the face during stream capture, achieving a more intuitive face registration processing.
By driving and guiding sounds on posture, users can more easily understand and perform face movements, improving the intuitiveness and accuracy of face registration, reducing misoperation and enhancing the accuracy of face recognition.
Smart Images

Figure CN118922870A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to a mobile object, a control device, and a control method, and more particularly, to a mobile object, a control device, and a control method capable of performing intuitive face registration. Background Art
[0002] In recent years, in various electronic devices such as smartphones, a face authentication method has been widely used, in which face authentication is performed based on face information of a user registered in advance, and when the face is recognized as the user's own face, the lock is released.
[0003] For example, Patent Document 1 discloses a face recognition device that recognizes an input face image by evaluating similarity with a registered face image registered in a registered face group selected based on an input result of an image or sound, and confirms that the recognized input face image is the person of the registered face image.
[0004] Reference List
[0005] Patent Literature
[0006] Patent Document 1: Japanese Patent Application Laid-Open No. 2004-302644 Summary of the invention Technical issues
[0007] In the related art, in the face registration process in which the face information of the user is registered in advance, the direction of the user's face is guided by using characters and images displayed on a display, but it is necessary to perform more intuitive face registration.
[0008] The present disclosure is made in consideration of such circumstances, so that intuitive face registration can be performed. Solution to the problem
[0009] According to one aspect of the present disclosure, a moving body and a control device include: a posture control part, which controls a posture drive for representing the movement of the user's face during stream capture through postures in a subsequent process of a tutorial for a face registration process in which the user's face is pre-registered; and a guide sound control part, which controls the output of a posture guide sound corresponding to the posture accompanying the posture drive.
[0010] A control method according to one aspect of the present disclosure includes: controlling a gesture drive that represents movement of a user's face during stream capture through gestures in a subsequent process of a tutorial of a face registration process that pre-registers a user's face; and controlling output of a gesture guide sound corresponding to a gesture accompanying the gesture drive.
[0011] In one aspect of the present disclosure, in a subsequent process of a tutorial of a face registration process for pre-registering a user's face, a gesture drive is controlled to represent the movement of a user's face during stream capture through gestures; and output of a gesture guide sound corresponding to the gesture accompanying the gesture drive is controlled. BRIEF DESCRIPTION OF THE DRAWINGS
[0012] [ Figure 1 ] is a diagram showing an example of a usage state of a mobile object to which the present technology is applied.
[0013] [ Figure 2 ] is a diagram for describing examples of gesture guiding sound and gesture driving.
[0014] [ Figure 3 ] is a block diagram showing a configuration example of an embodiment of a moving body.
[0015] [ Figure 4 ] is a diagram used to describe eigenvectors and center vectors.
[0016] [ Figure 5 ] is a diagram used to describe the threshold value.
[0017] [ Figure 6 ] is a flowchart for describing the face registration process.
[0018] [ Figure 7 ] is a block diagram showing an example of the configuration of an embodiment of a computer to which the present technology is applied. DETAILED DESCRIPTION
[0019] Hereinafter, specific embodiments to which the present technology is applied will be described in detail with reference to the accompanying drawings.
[0020] <Example of use of mobile body>
[0021] Figure 1 It is a diagram for describing a usage state of a mobile object to which the present technology is applied. Figure 1 An example of a state in which a user uses the mobile body 11 placed on a table as viewed from a lateral direction is shown.
[0022] For example, the mobile body 11 is an agent type robot device capable of performing autonomous travel and enables communication with the user to be more naturally and efficiently achieved. In addition, the mobile body 11 is, for example, a small robot configured with a certain size and weight so that the user can easily lift it with one hand.
[0023] The mobile body 11 has a configuration in which a hemispherical face portion 13 is provided on an upper portion of a main body 12 of an oblong shape in the longitudinal direction, and a hemispherical face portion 13 is provided on the front side (toward the Figure 1A camera 14 and an eye 15 are provided on one side in the right direction of the body 12, and a tire 16 is provided on the bottom surface of the body 12.
[0024] The face portion 13 is configured to be able to freely change direction in the up-down direction and the left-right direction by a driving mechanism built in the moving body 11 .
[0025] The camera 14 captures images in a direction toward the front of the face 13, and acquires still images or moving images. Figure 1 As shown in , when the user faces the face 13, the camera 14 can obtain a face image by capturing the user's face.
[0026] The eye 15 includes, for example, an LED (light emitting diode), an organic EL (electroluminescence), etc., and can express sight, blinking, etc. Figure 1 In FIG. 1 , only one eye 15 is shown, but as Figure 2 As shown in FIG. 1 , two eyes 15L and 15R are provided side by side on the face 13 when viewed from the front.
[0027] The tire 16 can be freely rotated by a driving mechanism built in the moving body 11 , and realizes moving operations of the moving body 11 , such as forward movement, backward movement, turning, and rotation.
[0028] The mobile body 11 configured as described above can register the face information of the user in the face database in advance, and can realize communication suitable for each user by performing face authentication processing when the user uses it.
[0029] In the case where the face authentication process is performed, the mobile body 11 outputs a guiding voice such as "Look at me" from a speaker (not shown). In response to this, when the user brings the face closer to the face portion 13 of the mobile body 11, the mobile body 11 acquires face information from a face image obtained by capturing the face of the user by the camera 14. Then, the mobile body 11 is able to recognize the face of each user by performing the face authentication process that evaluates the similarity between the face information of the user and each of the plurality of face information registered in the face database.
[0030] For example, in the face authentication process using the feature vector as the face information, the feature vector obtained from the face of the user captured by the camera 14 is used as the recognition target, and the distance (cosine distance (similarity), Euclidean distance, etc.) between each recognition target feature vector and each registered feature vector is calculated on a one-to-one basis. Then, in the face authentication process, it can be recognized that the face of the registered feature vector whose distance is equal to or greater than a predetermined threshold and the face of the user captured by the camera 14 are the same person.
[0031] By the way, in the face registration process of registering the face information of the user in advance, the user must be guided to change the direction of the face in the forward, right, left, upward and downward directions toward the camera. For example, when the face registration process is performed in a smart phone, the user is guided to change the direction of the face by using characters and images displayed on the display.
[0032] On the other hand, the mobile body 11 is configured to perform an intuitive face registration process without using a display, that is, the user can easily understand the face registration process in which the user changes the direction of the face by using a gesture guidance voice and a gesture-driven tutorial.
[0033] Reference Figure 2 , an example of a gesture guide sound and a gesture drive in a tutorial when the mobile body 11 performs a face registration process will be described.
[0034] For example, gesture driving expresses the movement (speed and direction) of the user's face at the time of stream capture through the gesture of the face 13 of the moving body 11, and the gesture guide sound outputs a constant rhythm corresponding to the speed of the face 13 of the moving body 11 according to the gesture.
[0035] First, if Figure 2 As shown in A of , the moving body 11 outputs a preliminary guidance voice “Move your face slowly according to the voice” with the face 13 facing the front, explains how to move the face, and then starts the tutorial.
[0036] For example, in the tutorial, Figure 2 As shown in FIG. 1B , the moving body 11 outputs a gesture guidance sound “one, two, three, four, five!” with a constant rhythm, and also performs gesture driving to rotate the face 13 at a constant speed so as to face the right side. Similarly, as Figure 2 As shown in B of FIG. 1 , the moving body 11 outputs a gesture guiding sound “One, two, three, four, five!” with a constant rhythm, and also performs gesture driving to rotate the face 13 at a constant speed so as to face the left side.
[0037] In addition, if Figure 2 As shown in FIG. 1D , the moving body 11 outputs a gesture guidance sound “one, two, three, four, five!” with a constant rhythm, and also performs gesture driving to rotate the face 13 at a constant speed so as to face upward. Similarly, as Figure 2 As shown in E of FIG. , the moving body 11 outputs a gesture guiding sound “One, two, three, four, five!” with a constant rhythm, and also performs gesture driving to rotate the face 13 at a constant speed so as to face downward.
[0038] After the tutorial, refer to Figure 6As described in the flowchart, the mobile body 11 starts streaming capture, and similar to the gesture guide sound at the time of gesture driving, a face direction guide sound "one, two, three, four, five!" with a constant rhythm is output, and the face direction guide sound guides the movement of the user's face. At this time, the mobile body 11 performs streaming capture of the user's face while the face 13 is fixed to the front, and guides the user to change the direction of the face in the right direction, the left direction, the upward direction, and the downward direction in sequence. In addition, while changing the direction of the face in the right direction, the left direction, the upward direction, and the downward direction, respectively, the mobile body 11 guides the user so that the user's face is always facing the front.
[0039] It should be noted that in the tutorial, not all gestures are necessarily performed in the four directions of right, left, upward, and downward, and only gestures need to be performed in at least one direction. For example, the mobile body 11 may perform gestures in one direction of the left and right directions and one direction of the upward and downward directions in the tutorial.
[0040] <Configuration example of mobile body>
[0041] Figure 3 is a block diagram showing a configuration example of an embodiment of a moving body.
[0042] like Figure 3 As shown in FIG. 1 , the mobile body 11 includes a sound output portion 21, a drive portion 22, an image capture portion 23, a storage portion 24, a face registration processing portion 25, and a threshold setting portion 26. The face registration processing portion 25 performs face registration processing and includes a guide sound control portion 31, a posture control portion 32, a feature vector extraction portion 33, and a center vector calculation portion 34.
[0043] The sound output section 21 includes, for example, a speaker or the like, and outputs a guide sound necessary for guidance when the face registration process is performed according to the control of the guide sound control section 31 .
[0044] The driving section 22 includes, for example, a motor and the like, and performs a posture drive in which the face 13 is moved as shown in FIG. Figure 2 The rotation according to the control of the posture control section 32 moves the face 13 so as to face the right direction, the left direction, the upward direction, and the downward direction, respectively.
[0045] The image capture part 23 includes, for example, an imaging element included in the camera 14, is capable of acquiring an image by capturing an image of an object in front of the face 13, acquiring a face image by, for example, stream capturing the user's face, and supplying the face image to the feature vector extraction part 33.
[0046] The storage section 24 includes, for example, a nonvolatile memory such as a flash memory, and registers the center vector calculated by the center vector calculation section 34 in the face registration process in the face database.
[0047] The threshold value setting section 26 sets a threshold value used in the face authentication process for evaluating the similarity with the center vector registered in the face database, and stores the threshold value in the storage section 24. Figure 5 The threshold value set by the threshold value setting section 26 is described.
[0048] During the tutorial, the guidance sound control section 31 controls the output of the gesture guidance sound according to the gesture of the face 13 of the mobile body 11, that is, as shown in FIG. Figure 2 The output of the preliminary guidance sound and the posture guidance sound. In addition, the guidance sound control section 31 controls as will be described later with reference to Figure 6 The output of the start guide sound, the face direction guide sound, the end guide sound, etc. as described in the flowchart is performed, and the sound output part 21 outputs the guide sound.
[0049] During the tutorial, the gesture control section 32 controls the movement of the user's face at the time of streaming capture, that is, gesture driving that indicates the speed and direction in which the user moves the face through the gesture of the face 13 of the moving body 11. That is, as described with reference to FIG. Figure 2 As described above, the gesture control section 32 controls the drive section 22 to perform gesture driving for rotating the face 13 toward the right, left, upward, and downward directions at a constant speed according to the rhythm of the gesture guide sound.
[0050] The feature vector extraction section 33 extracts a plurality of feature vectors from the face images of various angles acquired by the stream capture of the image capture section 23 , and supplies the feature vectors to the center vector calculation section 34 .
[0051] The central vector calculation section 34 calculates a central vector which is the center of all the feature vectors supplied from the central vector calculation section 34 .
[0052] Here, we will refer to Figure 4 Describe the eigenvectors and the center vector.
[0053] As described above, the mobile body 11 performs stream capture in the face registration process, and the feature vector extraction section 33 extracts the plurality of feature vectors from the face images of various angles acquired by the stream capture. Figure 4 An image of the plurality of eigenvectors is shown, but in fact, the eigenvectors are vectors on a 512-dimensional hypersphere.
[0054] Then, the center vector calculation section 34 calculates the center vector using all the face images acquired by the stream capture. That is, the result is stabilized by using the center of the feature vector extracted from the face images at various angles. It should be noted that the center vector calculation section 34 may calculate the center vector at a time point when a predetermined number (e.g., 50) of feature vectors have been accumulated. In addition, the feature vector extraction section 33 is pre-learned so that the features of the face of the same person are close and the features of the faces of different people are far apart.
[0055] In addition, if Figure 5 As shown in FIG. 1 , the threshold value setting section 26 sets the target threshold value D θ , maximum distance threshold D R And the target threshold D θ and the maximum distance threshold D R Any of the maximum values in is set as the threshold to be used for face authentication processing. For example, the target threshold D θ is a value determined when designing the moving body 11, and the maximum distance threshold D R It is the value corresponding to the distance to the eigenvector at the farthest position centered on the center vector.
[0056] By using the threshold value set in this manner, the mobile body 11 can relatively easily verify and implement the face authentication process, and the processing load can be reduced.
[0057] When the face authentication process is performed, the mobile body 11 calculates the distance between the feature vector extracted by the feature vector extraction section 33 and the registered center vector, and sets the center vectors whose distance is shortest and falls within the range of the threshold set by the threshold setting section 26 to be the same.
[0058] Here, the mobile body 11 is equipped with a feature extractor that is learned to minimize the distance between the central vector or representative vector of each face class and the feature vector of the same face class. That is, the feature extractor distributes the feature vectors of the face of a certain person at various angles so as to be dispersed around the central vector or representative vector. Therefore, the center of the feature vector of the face image at various angles acquired by stream capture can capture the approximate center of the distribution of the feature vector of the face. In addition, since the feature vector output from the feature extractor is normalized and the constraint that the feature vector exists on a hypersphere is added, it is possible to avoid being affected by the length of the vector, and therefore, it is expected that the center of the feature vector of the face image collected by stream capture is stable. On the other hand, in a feature extractor that is learned not to minimize the distance between the central vector or representative vector of each face class and the feature vector of the same face class but to simply make the distance between the feature vectors of the same face closer and the distance between the feature vectors of different faces farther, it is not guaranteed that the shape of the distribution is dispersed in a concentric manner (a hypersphere rather than a circle because it is multidimensional), and it is not expected that the accuracy of face recognition can be performed using the center of the feature vector of the face image collected by stream capture.
[0059] <Processing Example of Face Registration Processing>
[0060] Will refer to Figure 6 The face registration processing performed by the face registration processing section 25 is described with reference to the flowchart shown in FIG.
[0061] For example, when the user uses the mobile body 11 for the first time and says “Let's be friends”, the face registration process starts as a result of the voice recognition process of the utterance.
[0062] In step S11, the guidance sound control part 31 controls the output of the preliminary guidance sound, and causes the sound output part 21 to output the preliminary guidance sound. For example, the sound output part 21 outputs the preliminary guidance sound "Let's take a moment to recognize your face to commemorate our meeting", and outputs the preliminary guidance sound "I learn at various angles". Therefore, it will be explained to the user to capture the face at the multiple angles. When the user is the owner of the mobile body 11, the user's name is registered in advance, and the user's name can be confirmed at the beginning of the face registration process. Then, the sound output part 21 outputs the preliminary guidance sound "I will demonstrate it first" to explain to the user that the tutorial starts, and the process proceeds to step S12.
[0063] In step S12, the guidance sound control section 31 controls the output of the gesture guidance sound, and the gesture control section 32 controls the gesture driving. Figure 2 As described above, the tutorial is performed by the driving section 22 performing gesture driving while the sound output section 21 outputs the gesture guiding sound.
[0064] In step S13, the guide sound control section 31 controls the output of the start guide sound, and causes the sound output section 21 to output the start guide sound. For example, the sound output section 21 outputs the start guide sound "Do you understand? Move your face as slowly as possible like this" to explain how to move the face, and outputs the start guide sound "Okay, now I will start to remember your face" to announce that the capture of the face is about to start. Then, the sound output section 21 outputs the start guide sound "Stare at my face" so that the user faces the front, and then the process proceeds to step S14.
[0065] In step S14 , stream capturing by the image capturing section 23 starts, and face images are sequentially supplied from the image capturing section 23 to the feature vector extracting section 33 .
[0066] In step S15, the guide sound control section 31 controls the output of the face direction guide sound, and starts outputting the face direction guide sound from the sound output section 21. Therefore, the sound output section 21 starts outputting the face direction guide sound, for example, such as "From there, turn right. One, two, three, four, five!", "Look at my face again. Now, turn left. One, two, three, four, five!", "Look at my face again. Now, look up. One, two, three, four, five!", and "Look at my face again. Now, look down. One, two, three, four, five!".
[0067] In step S16, the feature vector extraction section 33 detects the face of the user from the face image supplied from the image capture section 23. Here, in the feature vector extraction section 33, when the size of the detected face of the user is small, it is difficult to detect the part point in step S17, and therefore, it does not proceed to step S17 until the face of the user having a predetermined size or larger is detected. For example, in this case, by driving the tires 16, the moving body 11 moves so that a face image of an appropriate size can be captured.
[0068] In step S17, the feature vector extraction section 33 detects part points (part points are features of corresponding parts such as eyes, nose and mouth) from the face of the user detected in step S16, and estimates the direction (yaw, pitch, roll) of the face based on the part points. Here, when the estimated face direction is outside the specified range, or when the face direction cannot be estimated, the feature vector extraction section 33 does not proceed to step S18.
[0069] In step S18 , the feature vector extraction section 33 adjusts the position using the part points estimated in step S17 , extracts the feature vector of the user's face, and supplies the feature vector to the center vector calculation section 34 .
[0070] In step S19, the face registration processing section 25 determines whether the face direction guidance is finished. For example, when the output of the face direction guidance sound started in step S15 is finished, that is, when all the guidance of directing the user's face in the right direction, the left direction, the upward direction, and the downward direction is performed, the face registration processing section 25 determines that the face direction guidance is finished.
[0071] In step S19, when the face registration processing part 25 determines that the face direction guidance is not finished, the process returns to step S16, and the same process is repeatedly performed thereafter. On the other hand, in step S19, when the face registration processing part 25 determines that the face direction guidance is finished, the process proceeds to step S20.
[0072] In step S20, the stream capturing of the image capturing section 23 ends, and the supply of the face image from the image capturing section 23 to the feature vector extracting section 33 stops. At this time, the center vector calculating section 34 accumulates the plurality of feature vectors supplied from the feature vector extracting section 33 in the period of time during which the stream capturing is performed.
[0073] In step S21 , the center vector calculation section 34 obtains a center vector by calculating the center of the plurality of feature vectors supplied from the feature vector extraction section 33 , and registers the center vector in the face database of the storage section 24 .
[0074] In step S22, the guidance sound control part 31 controls the output of the end guidance sound, and causes the sound output part 21 to output the end guidance sound. Here, the processing of step S22 can be performed at the time required to perform the processing of steps S20 and S21. For example, at the time required to perform the processing of steps S20 and S21, the sound output part 21 outputs the end guidance sound, such as "Okay, so I won't forget your face, so wait a moment", or outputs the end guidance sound, such as "I'm remembering you now, so wait a moment". When the user uses the mobile body 11 for the first time, the user's name can be registered. Then, when the processing of steps S20 and S21 ends, the sound output part 21 outputs the end guidance sound "I remember!", and the processing ends.
[0075] As described above, the mobile body 11 can complete the registration of the user's face information by guiding sound and gesture driving without using a display. That is, in the tutorial, the mobile body 11 performs gesture driving that expresses the movement of the user's face at the time of stream capture by gestures based on the speed and direction (movement range of the face) of the face 13 of the mobile body 11, and outputs a gesture guiding sound at a constant rhythm corresponding to the speed of the face 13 of the mobile body 11 according to the gesture, so that the user can easily grasp the way to move the face at the time of stream capture. Therefore, the user can move the direction of the face according to the face direction guiding sound without being confused.
[0076] In addition, in the face registration process performed by stream capture, the mobile body 11 can avoid, for example, the feature vector when facing a direction other than the frontal face is different from the feature vector of the frontal face, causing the feature vector to be regarded as a different person. In addition, the mobile body 11 can avoid, for example, the face registration process is not completed due to the fact that the direction is not detected by the estimation accuracy of the face angle, and the capture is repeated many times until the face registration process ends. Then, by calculating the center vector that is the center of the multiple feature vectors extracted from the face images at various angles and registering it in the face database, the mobile body 11 can provide a more robust and very accurate face recognition function. That is, by using the face images at various angles obtained by stream capture, the mobile body 11 can improve the face recognition accuracy.
[0077] <Example of computer configuration>
[0078] Figure 7 : is a block diagram showing an example of the configuration of hardware of a computer that executes the above-described series of processes using a program.
[0079] In the computer, a CPU (Central Processing Unit) 101, a ROM (Read Only Memory) 102, a RAM (Random Access Memory) 103, and an EEPROM (Electrically Erasable Programmable Read Only Memory) 104 are interconnected through a bus 105. In addition, an input / output interface 106 is connected to the bus 105, and the input / output interface 106 is connected to the outside. It should be noted that the process of extracting feature values can be performed by a GPU (Graphics Processing Unit), a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array), etc. as well as the CPU.
[0080] In the computer configured as described above, the CPU 101 loads the program stored in, for example, the ROM 102 and the EEPROM 104 into the RAM 103 via the bus 105, and executes the program, thereby performing the above-mentioned series of processing. The program executed by the computer (CPU 101) may be written in advance in the ROM 102, or may be installed or updated in the EEPROM 104 from the outside via the input / output interface 106.
[0081] Here, the processing performed by the computer according to the program in this specification is not necessarily performed in time series according to the order described in the flowchart. In other words, the processing performed by the computer according to the program includes processing performed in parallel or individually (for example, parallel processing or object processing).
[0082] In addition, the program may be processed by a single computer (processor), or may be processed by a plurality of computers in a distributed manner. In addition, the program may be transferred to a remote computer so as to be executed by the remote computer.
[0083] In addition, in this specification, a system refers to a collection of multiple components such as devices and modules (parts), and it is not important whether all components are located in the same housing. Therefore, multiple devices housed in separate housings and connected to each other via a network and a device in which multiple modules are housed in one housing are both systems.
[0084] In addition, for example, the configuration described as one device (or processing section) may be divided and configured as the plurality of devices (or processing sections). Conversely, the configuration described above as a plurality of devices (or processing sections) may be configured together as one device (or processing section). In addition, a configuration other than the above configuration may be added to the configuration of each device (or each processing section). In addition, if the configuration and operation of the entire system are substantially the same, a portion of the configuration of a certain device (or processing section) may be included in the configuration of another device (or another processing section).
[0085] For example, the present technology may also have a configuration of cloud computing in which one function is shared by the plurality of devices via a network so as to be processed in a collaborative manner.
[0086] In addition, for example, the above-mentioned program can be executed in any device. In this case, the device may have necessary functions (such as functional blocks) so that necessary information can be obtained.
[0087] In addition, the corresponding steps described using the above flowchart may be performed by one device, or may be shared and performed by the plurality of devices. In addition, when one step includes the plurality of processes, the plurality of processes included in the one step may be performed by one device, or may be shared and performed by the plurality of devices. In other words, the plurality of processes included in one step may be performed as a plurality of steps. Conversely, the processes described as the plurality of steps may be performed together as one step.
[0088] It should be noted that, with respect to a program executed by a computer, the steps describing the program may be executed in chronological order according to the order described in this specification, or may be executed individually in parallel or at necessary timing (such as when a call is made). That is, as long as there is no inconsistency, the processing of the corresponding steps may be executed in an order different from the above order. In addition, the processing of the steps describing the program may be executed in parallel with the processing of another program, or may be executed in combination with the processing of another program.
[0089] It should be noted that, as long as there is no inconsistency, the present technology described in this specification can be implemented independently. It should be understood that any multiple present technologies can be used in combination. For example, some or all of the present technologies described in any embodiment can be implemented in combination with some or all of the present technologies described in other embodiments. In addition, some or all of the present technologies in any of the above-mentioned present technologies can be implemented in combination with other technologies not described above.
[0090] <Example of configuration combination>
[0091] The present technology may also have the following structures.
[0092] (1) A mobile object, comprising:
[0093] a gesture control section that controls gesture driving for expressing movement of the user's face at the time of stream capture by gestures in a subsequent process of a tutorial of a face registration process for registering the user's face in advance; and
[0094] The guidance sound control section controls output of a gesture guidance sound corresponding to a gesture accompanying the gesture driving.
[0095] (2) The mobile object according to (1), wherein the gesture control section indicates, by a gesture of the face of the mobile object, a speed and a direction in which the user moves the face at the time of stream capturing.
[0096] (3) The moving body according to (2), wherein the guide sound control section outputs a constant rhythm corresponding to the speed as the gesture guide sound.
[0097] (4) The mobile body as described in any one of (1) to (3), wherein the guide sound control part controls the output of a face direction guide sound that guides the movement of the user's face in the streaming capture performed after the tutorial.
[0098] (5) The moving object according to (4), further comprising:
[0099] a feature vector extraction section that extracts a plurality of feature vectors from a user's face image at various angles acquired by changing the direction of the user's face according to the guidance of the face direction guiding sound when streaming capture is performed; and
[0100] The center vector calculation section calculates a center vector which is the center of the plurality of feature vectors and registers the center vector in the face database.
[0101] (6) The moving object according to (5), further comprising:
[0102] The threshold setting section sets a threshold used in a face authentication process for evaluating similarity with a center vector registered in a face database.
[0103] (7) A moving body as described in (6), wherein the threshold setting part sets any one of a first threshold, a second threshold and a third threshold as the threshold, the first threshold is a value determined at the time of design, the second threshold corresponds to the distance between the center vector and the feature vector at the farthest position, and the third threshold is the maximum value of the first threshold and the second threshold.
[0104] (8) A control device comprising:
[0105] a gesture control section that controls gesture driving for expressing movement of the user's face at the time of stream capture by gestures in a subsequent process of a tutorial of a face registration process for registering the user's face in advance; and
[0106] The guidance sound control section controls output of a gesture guidance sound corresponding to a gesture accompanying the gesture driving.
[0107] (9) A control method comprising:
[0108] Through the control device,
[0109] controlling a gesture drive that expresses movement of the user's face at the time of stream capture by gestures in a subsequent process of a tutorial of a face registration process that registers the user's face in advance; and
[0110] The output of the gesture guidance sound corresponding to the gesture accompanying the gesture driving is controlled.
[0111] It should be noted that the present embodiment is not limited to the above-described embodiment, and various modifications may be made without departing from the scope of the present disclosure. It should be noted that the effects described in this specification are not restrictive but merely illustrative, and other effects may be provided.
[0112] Label list
[0113] 11 Moving Object
[0114] 12 Main Body
[0115] 13. Face
[0116] 14 Camera
[0117] 15 Eyes
[0118] 16 Tires
[0119] 21 Sound output section
[0120] 22 drive part
[0121] 23 Image Capture Section
[0122] 24 Storage section
[0123] 25Face registration processing part
[0124] 26Threshold Setting Section
[0125] 31 Guide sound control part
[0126] 32 Posture Control Section
[0127] 33 Feature vector extraction part
[0128] 34 Center vector calculation part
Claims
1. A mobile object, comprising: a gesture control section that controls gesture driving for expressing movement of the user's face at the time of stream capture by gestures in a subsequent process of a tutorial of a face registration process for registering the user's face in advance; and The guidance sound control section controls output of a gesture guidance sound corresponding to a gesture accompanying the gesture driving.
2. The mobile body according to claim 1, wherein the gesture control section indicates, by a gesture of the face of the mobile body, a speed and a direction in which the user moves the face at the time of stream capturing.
3. The moving body according to claim 2, wherein the guide sound control section outputs a constant rhythm corresponding to the speed as the gesture guide sound.
4. The mobile body according to claim 1, wherein the guide sound control section controls output of a face direction guide sound that guides movement of the user's face in stream capturing performed after a tutorial.
5. The mobile body according to claim 4, further comprising: a feature vector extraction section that extracts a plurality of feature vectors from a user's face image at various angles acquired by changing the direction of the user's face according to the guidance of the face direction guiding sound when streaming capture is performed; and The center vector calculation section calculates a center vector which is the center of the plurality of feature vectors and registers the center vector in the face database.
6. The mobile body according to claim 5, further comprising: The threshold setting section sets a threshold used in a face authentication process for evaluating similarity with a center vector registered in a face database.
7. A moving body as described in claim 6, wherein the threshold setting part sets any one of a first threshold, a second threshold and a third threshold as the threshold, the first threshold is a value determined at the time of design, the second threshold corresponds to the distance to the feature vector at the farthest position centered on the center vector, and the third threshold is the maximum value of the first threshold and the second threshold.
8. A control device comprising: a gesture control section that controls gesture driving for expressing movement of the user's face at the time of stream capture by gestures in a subsequent process of a tutorial of a face registration process for registering the user's face in advance; and The guidance sound control section controls output of a gesture guidance sound corresponding to a gesture accompanying the gesture driving.
9. A control method, comprising: Through the control device, controlling a gesture drive that expresses movement of the user's face at the time of stream capture by gestures in a subsequent process of a tutorial of a face registration process that registers the user's face in advance; and The output of the gesture guidance sound corresponding to the gesture accompanying the gesture driving is controlled.
