JP2014014899A - Information processing apparatus, information processing system, service provider system, and information processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information processing apparatus, an information processing system, a service provider system, and an information processing method that are able to make a switchover to an appropriate service depending on the circumstance of a target that receives the service.SOLUTION: An information processing apparatus according to an embodiment includes an acquisition unit, a first detection unit, a second detection unit, a selection unit, and a first determination unit. The acquisition unit acquires motion vectors of a plurality of individual targets. The first detection unit detects a first group indicating a collection of motion vectors that are equal to or larger than a threshold value and are similar to one another. The second detection unit detects a second group indicating a collection of motion vectors that are smaller than the threshold value. The selection unit selects one of the first group and the second group. The first determination unit decides to provide a first service if the selection unit selects the first group, and decides to provide a second service different from the first service if the selection unit selects the second group.

Description

  Embodiments described herein relate generally to an information processing apparatus, an information processing system, a service providing system, and an information processing method.

  2. Description of the Related Art Conventionally, service providing apparatuses that perform services such as provision of product information indicating information such as products and advertisements in public spaces (such as shopping malls and on the street) are known. For example, a form in which a display device including a display for displaying product information is used as a service providing device is known. In this form, interactive information is provided (interactive service is performed) by combining with a sensor that detects an object (for example, a person) to provide a service, taking advantage of the property that display contents can be changed in real time. You can also. Also, in order to pursue the interactivity of the service to be provided, a form in which a robot apparatus that performs services such as notification of product information is used as a service providing apparatus is also known.

  Here, for example, when a plurality of persons are gathered and approaching the service providing apparatus, it can be estimated that a group that performs actions is approaching the service providing apparatus. For a group consisting of a set of people who have established a relationship such as mutual communication, by providing a service with close conversation (providing interactive information), It is thought that all the persons who belong can be satisfied.

  On the other hand, for example, when a large number of people surround the service providing apparatus, there may be a person who does not have a relationship such as communication. In such a case, it is not appropriate to perform a service based on close conversation by regarding a plurality of persons surrounding the service providing apparatus as one group. In such cases, acting and descriptive services, that is, non-interactive services, can work uniformly on each person who provides the service, so the effect of providing the service Is big.

  However, in the past, whether the target to provide the service is a set of persons who have established a relationship such as communicating with each other and whether the relationship such as communicating with each other has been established. There is no technology for switching the service to be provided to an appropriate one when there is a group of unknown persons.

JP 2011-194507 A JP 2005-347321 A JP 2008-213082 A International Publication No. WO2006 / 085383

  The problem to be solved by the present invention is to provide an information processing apparatus, an information processing system, a service providing system, and an information processing method that can be switched to an appropriate service according to the situation of an object to provide the service.

  The information processing apparatus according to the embodiment includes an acquisition unit, a first detection unit, a second detection unit, a selection unit, and a first determination unit. The acquisition unit acquires each movement vector of the plurality of objects. The first detection unit detects a first group indicating a set of movement vectors similar in size to each other that have a size equal to or greater than a threshold value. The second detection unit detects a second group indicating a set of movement vectors that are less than the threshold. The selection unit selects either the first group or the second group. The first determination unit determines to perform the first service when the selection unit selects the first group, and when the selection unit selects the second group, the second determination unit is different from the first service. Decide to perform the service.

  An information processing system according to an embodiment is an information processing system including an information processing device and one or more server devices, and includes an acquisition unit, a first detection unit, a second detection unit, a selection unit, and a first determination unit. Prepare. The acquisition unit acquires each movement vector of the plurality of objects. The first detection unit detects a first group indicating a set of movement vectors similar in size to each other that have a size equal to or greater than a threshold value. The second detection unit detects a second group indicating a set of movement vectors that are less than the threshold. The selection unit selects either the first group or the second group. The first determination unit determines to perform the first service when the selection unit selects the first group, and when the selection unit selects the second group, the second determination unit is different from the first service. Decide to perform the service.

  The service providing system according to the embodiment is a service providing system including one or more server devices and an output device, and the one or more server devices include an acquisition unit, a first detection unit, a second detection unit, and a selection unit. A determination unit. The acquisition unit acquires each movement vector of the plurality of objects. The second detection unit detects a second group indicating a set of movement vectors that are less than the threshold. The selection unit selects either the first group or the second group. The determination unit determines to perform the first service when the selection unit selects the first group, and determines the second service different from the first service when the selection unit selects the second group. Decide what to do. On the other hand, the output device includes at least an output unit that performs output according to the determination by the determination unit.

  The information processing method of the embodiment includes an acquisition step, a first detection step, a second detection step, a selection step, and a first determination step. In the acquisition step, each movement vector of the plurality of objects is acquired. In the first detection step, a first group indicating a set of movement vectors having magnitudes equal to or larger than a threshold and similar to each other is detected. The second detection step detects a second group indicating a set of movement vectors that are less than the threshold value. The selection step selects either the first group or the second group. The first determining step determines to perform the first service when the first group is selected, and determines to perform the second service different from the first service when the second group is selected. To do.

The figure which shows the example of an external appearance of the robot apparatus of embodiment. The figure which shows the hardware structural example of the robot apparatus of embodiment. The figure which shows the function structural example of the robot apparatus of embodiment. The figure which shows an example of the group detected of embodiment. The figure which shows the example of the data memorize | stored in the person position memory | storage part of embodiment. The figure which shows the example of the data memorize | stored in the voice call content storage part of embodiment. The figure which shows the example of the data in the application memory | storage part classified by use of embodiment. 6 is a flowchart illustrating an example of a processing operation performed by the robot apparatus according to the embodiment.

  Hereinafter, embodiments of an information processing apparatus, an information processing system, a service providing system, and an information processing method according to the present invention will be described in detail with reference to the accompanying drawings.

(First embodiment)
In the first embodiment, it is assumed that the robot apparatus is used as a service providing apparatus. FIG. 1 is a diagram showing an example of the appearance of the robot apparatus 1 of the present embodiment. As shown in FIG. 1, the robot apparatus 1 is configured by a robot imitating a human shape, and includes a head 10, a right arm 20, a left arm 30, a body 40, a right leg 50, and a left. Leg portion 60. In addition, the robot apparatus 1 includes a first camera 101, a second camera 102, a speaker 161, and a microphone 162. The first camera 101 is disposed on the head 10, and the second camera 102, the speaker 161, and the microphone 162 are disposed on the body portion 40.

  Although not shown in detail, the head 10 is driven by the motor 171T for driving the head 10 in the tilt direction (up and down direction) and the head 10 in the Pan direction (left and right direction). A motor 172P is disposed. The position (imaging position) of the first camera 101 provided on the head 10 can be changed by rotating the motors 171T and 172P. The body portion 40 is provided with a motor 177W for driving the body portion 40. In this example, the second camera 102 provided in the body 40 has a wider-angle camera lens than the first camera 101 provided in the head 10 and can capture a wider range of objects (for example, a person). Is possible. Further, in this example, a drive unit (motor) for driving the leg part is not arranged on the leg part (50, 60) of the robot. That is, the robot in this example is a stationary robot.

  FIG. 2 is a diagram illustrating a hardware configuration example of the robot apparatus 1. As shown in FIG. 2, the robot apparatus 1 includes at least an information processing apparatus 100, a first camera 101, a second camera 102, a speaker 161, a microphone 162, and a driving unit 170. The information processing apparatus 100 includes a CPU 181, a memory 182, an HDD 183, an external I / O 184, and a communication unit 185. The CPU 181 controls the overall operation of the robot apparatus 1 by executing a program stored in the memory 182 or the HDD 183. The external I / O 184 inputs and outputs data related to the robot operation. The communication unit 185 transmits / receives data to / from the store terminal 200 via a wireless LAN or the like.

  In this example, the drive unit 170 includes the motors 171T, 172P, and 177W described above. Note that the hardware configuration of the robot apparatus 1 shown in FIG. 2 is an example, and for example, a configuration in which the communication unit 185 is not provided may be employed. In the example of FIG. 2, a speaker 161 and a driving unit 170 corresponding to an output unit 150 described later are provided separately from the information processing apparatus 100, but the output unit 150 is mounted on the information processing apparatus 100, for example. Form may be sufficient.

  FIG. 3 is a block diagram illustrating a functional configuration example of the robot apparatus 1. In the example of FIG. 3, the functions of the robot apparatus 1 include a person movement vector detection unit 110, an acquisition unit 117, a group detection / selection unit 120, an initial operation control unit 130, a first determination unit 140, A use-specific application storage unit 141, an application execution unit 142, and an output unit 150 are included. The specific contents of each function will be described below.

  The person movement vector detection unit 110 detects the position of the person around the robot apparatus 1 and detects the movement vector of the person from the time series change of the person position. In the present embodiment, the person movement vector detection unit 110 detects whether a person's face is facing the robot apparatus 1 and adds attribute information indicating attributes such as the gender and age of the person. It is also possible to manage the attribute information of each person. As shown in FIG. 3, the person movement vector detection unit 110 includes a face detection unit 111, an attribute determination unit 112, a person detection unit 113, a person position calculation unit 114, a movement vector detection unit 115, and a person position storage. Part 116.

  Each time shooting by the second camera 102 is performed, the person position calculation unit 114 acquires an image shot by the second camera 102 and calculates position information indicating the position of the person in the acquired image. As an example, the person position calculation unit 114 according to the present embodiment detects the position and size of the person's head in the image acquired from the second camera 102 using the head detection algorithm of the person detection unit 113. When a plurality of persons are shown in the image acquired from the second camera 102, the person detection unit 113 can detect the position and size of each head of the plurality of persons. The person position calculation unit 114 calculates the relative angle between the robot apparatus 1 and the person based on information such as the detected head size, the installation position of the camera (second camera 102 in this example), and the angle of view of the camera. The distance can be calculated. This calculation result is converted into an XY coordinate system set so that the ground is the reference plane and the position of the robot apparatus 1 is the origin, and is output (recorded) to the person position storage unit 116. The calculation result converted as described above becomes position information indicating the position of the person, and is recorded in the person position storage unit 116 in association with the person information for identifying the person.

  In the present embodiment, the position information is calculated using an image captured by the camera (second camera 102). However, any device other than the camera can be used as long as the position information can be calculated. May be used. For example, a form using a device such as a laser range finder, a human sensor, or a floor pressure sensor may be used. Moreover, the form using combining the above-mentioned device may be sufficient.

  The face detection unit 111 uses the first camera 101 to perform face detection on the position (person position) indicated by the position information calculated by the person position calculation unit 114. In the present embodiment, the first camera 101 is attached to a pan / tilt head whose camera angle can be controlled in the Tilt direction or the Pan direction by motors 171T and 172P. If the position of the person who wants to perform face detection is outside the range of the angle of view of the first camera 101, the face detection unit 111 controls the camera platform so that the position of the person falls within the angle of view of the camera. You can also For the face detection by the face detection unit 111, various known techniques can be used. When a face is detected by the face detection unit 111, the attribute determination unit 112 determines gender and age using a known attribute determination algorithm. Then, the attribute determination unit 112 records the attribute information indicating the gender and the age in the person position storage unit 116 in association with the position information indicating the position of the person whose face is detected.

  The movement vector detection unit 115 detects the movement vector of the person corresponding to the position indicated by the calculated position information every time the position information is calculated by the person position calculation unit 114. The movement vector detection unit 115 calculates the position information calculated at the previous position information calculation timing according to the time length from the previous position information calculation timing by the person position calculation unit 114 to the current position information calculation timing. For each position indicated by, a range in which the person corresponding to the position moves before the current position information calculation timing is estimated. Then, the movement vector detection unit 115 sets the person corresponding to the position closest to the estimated range among the positions indicated by the position information calculated at the current position information calculation timing, and the position from which the range is calculated. The corresponding person is regarded as the same person, the movement vector of the person is detected, and the detected movement vector and person information for identifying the person are associated and recorded in the person position storage unit 116.

  Note that the method for detecting the movement vector of the person is not limited to the above-described method, and various known techniques can be used. For example, the movement vector detection unit 115 detects the change in the feature point of the person corresponding to the position indicated by the position information calculated by the person position calculation unit 114, so that the position information calculated by the person position calculation unit 114 is obtained. A person corresponding to the indicated position can be identified, and a movement vector can be detected for each identified person. For example, the movement vector detection unit 115 uses the face information detected by the face detection unit 111 to identify a person corresponding to the position indicated by the position information calculated by the person position calculation unit 114, and identifies the identified person. The movement vector can also be detected for each.

  The person position storage unit 116 of the present embodiment stores data (referred to as “detection data”) in which position information, a movement vector, and attribute information are associated with each person (for each person information). The movement vector detection unit 115 detects a movement vector of a person corresponding to the position indicated by the calculated position information each time position information is calculated by the person position calculation unit 114, and the position information and the detected movement are detected. The vector is recorded in an area in the person position storage unit 116 where the detection data of the person is stored. In other words, the movement vector detection unit 115 associates the position information calculated by the person position calculation unit 114 and the movement vector detected by the person with the person information for identifying the person corresponding to the position indicated by the position information. It can also be understood that it is recorded in the person position storage unit 116. The data history of each person is held in the person position storage unit 116 until the person disappears from the camera detection range.

  The acquisition unit 117 acquires a movement vector of a person existing around the robot apparatus 1 (existing in the camera detection range). More specifically, the acquisition unit 117 accesses the person position storage unit 116 each time movement vector detection processing is performed by the movement vector detection unit 115, and acquires position information and movement vectors of each person at that time. To do.

  Based on the movement vector acquired by the acquisition unit 117, the group detection / selection unit 120 has a relationship such as mutual communication as shown in FIG. It is possible to detect a situation in which a set of persons assumed to be approaching the robot apparatus 1. Further, based on the movement vector acquired by the acquisition unit 117, the group detection / selection unit 120 is unknown whether a relationship such as mutual communication is established as shown in FIG. 4B. It is possible to detect a situation in which a group of people is surrounding the robot apparatus 1. The group detection / selection unit 120 has a function of selecting a group that can effectively provide a service.

  As illustrated in FIG. 3, the group detection / selection unit 120 includes a first detection unit 121, a second detection unit 122, and a selection unit 123. Each time the acquisition process by the acquisition unit 117 is performed, the first detection unit 121 has a plurality of movement vectors acquired by the acquisition unit 117, the magnitudes of which are equal to or larger than a threshold value and similar to each other. A first group indicating a set is detected. The first group can be regarded as a set of movement vectors of persons who have a relationship such as mutual communication. More specifically, it is as follows.

The first detection unit 121 selects one person Ha from a plurality of persons corresponding one-to-one with the plurality of movement vectors acquired by the acquisition unit 117, and selects the first person Ha for the selected person Ha. A first label number that can identify one group is assigned. The first label number is a number that is counted up from the initial value by the number of the first group. In this example, when the first label number is already assigned to the person Ha, a new first label number is not assigned to the person Ha. Next, the first detection unit 121 uses the movement vector of the person Hb existing within the range defined by the personal space (for example, within a radius of 1.2 m) around the position indicated by the position information of the selected person Ha. The similarity with the movement vector of the person Ha is determined. However, if the movement vector of the person Ha or Hb is less than the threshold (for example, less than 0.5 km / h), the determination is not performed, and the person whose movement vector is less than the threshold is excluded from the detection target. The similarity between the two movement vectors can be obtained by the following equation 1, where the larger movement amount is the vector x and the smaller movement vector is the vector y.

  If the similarity obtained by the above equation 1 is equal to or higher than a reference value (for example, 0.6 or higher), it is determined that they belong to the same first group, and the same first label number as the person Ha is assigned to the person Hb. In this example, when another first label number has already been assigned to the person Hb, the first label number assigned to the person Ha is assigned to the person Hb as the first label candidate number. Here, the first detection unit 121 assigns a first label number assigned to a person to whom a first label candidate number that is different from the assigned first label number is assigned for a predetermined time or longer (eg, 3 seconds or longer). The first label number assigned to the person is rewritten to the number indicated by the first label candidate number.

  Further, the first detection unit 121 sets the movement vector of the person Ha even when there is no other person within the range defined by the personal space around the position indicated by the position information of the selected person Ha. If it is equal to or greater than the threshold value, it is considered that the first group is composed only of the person Ha, and a first label number is assigned to the person Ha. The first detection unit 121 repeats the above labeling until there is no person to whom the first label number is not assigned. Thereby, a set of persons assigned with the same first label number is detected as a first group identified by the first label number. In other words, the first detection unit 121 selects a first group indicating a set of movement vectors that are similar in size to each other among the plurality of movement vectors acquired by the acquisition unit 117. Detected. In addition, the first detection unit 121 records the first label number or the first label candidate number in the person position storage unit 116 in association with the position information of the person to whom the first label number or the first label candidate number is assigned. When the labeling is completed, the first detection unit 121 includes information indicating the detected first group (for example, a first label number identifying the first group, personal information and detection data of a person included in the first group, etc. ) Is output to the selection unit 123.

  In this example, the first detection unit 121 determines that the set is the first when the size is equal to or larger than the threshold and the number of movement vectors belonging to a set of similar movement vectors is equal to or greater than the third threshold. Not detected as a group. More specifically, when the number of persons assigned the same first label number is equal to or greater than a third threshold (for example, 7 or more), the first detection unit 121 is a passerby who passes around the robot apparatus 1. Are not detected as the first group.

  Each time the acquisition process by the acquisition unit 117 is performed, the second detection unit 122 selects a second group indicating a set of movement vectors whose magnitude is less than the threshold among the plurality of movement vectors acquired by the acquisition unit 117. To detect. More specifically, the second detection unit 122 counts the number of persons whose faces are detected by the face detection unit 111 among a plurality of persons corresponding to a plurality of movement vectors whose sizes are less than the threshold. Is equal to or greater than the second threshold, a plurality of movement vectors having magnitudes less than the threshold are detected as the second group. Further details are as follows. It should be noted that the second group can be regarded as a set of movement vectors of persons whose unknown whether or not a relationship such as mutual communication is established.

  The second detection unit 122 has a range (for example, a radius of 2.4 m) defined around the position of the robot apparatus 1 from among a plurality of persons corresponding to the movement vectors acquired by the acquisition unit 117 on a one-to-one basis. A plurality of persons Hd that are within ± 50 cm) and whose movement vector is less than a reference value (for example, less than 0.5 km / h) are selected. Then, the second detection unit 122 counts the number of persons whose faces are detected among the plurality of persons Hd (the number of persons to which attribute information is added), and detects the faces among the plurality of persons Hd. If the number of persons who are present is equal to or more than the second threshold (for example, five or more), a second label number indicating the bystander group is assigned to each of the plurality of persons Hd and detected as the second group. In other words, the second detection unit 122 detects, as the second group, a set of a plurality of movement vectors whose sizes are less than the threshold among the plurality of movement vectors acquired by the acquisition unit 117. It can also be captured. The second detection unit 122 outputs information indicating the detected second group (for example, the second label number, personal information of persons included in the second group, detection data, and the like) to the selection unit 123.

  Note that the second detection unit 122 deletes the second label number for a person whose face is not detected or is not detected as the person Hd for a predetermined time or longer (for example, 3 seconds or longer). In this case, if the number of persons assigned with the second label number is less than a certain number, the second label number is deleted from all persons assigned with the second label number.

  In addition, the second detection unit 122 records the position information of the person to whom the second label number is assigned in the person position storage unit 116 in association with the second label number. As shown in FIG. 5, the person position storage unit 116 includes, for each person, a position locus storage area in which detection data of the person is stored, and a label number (first label number, first label candidate) assigned to the person. A label storage area in which label data capable of specifying at least one of the number and the second label number is stored. In the position trajectory storage area, detection data including position information, movement vectors, and attribute information at the time T is stored for each time T at which an image captured by the second camera 102 is acquired. In the label storage area, label data at the time T is stored for each time T at which an image captured by the second camera 102 is acquired.

  The selection unit 123 selects either the first group or the second group. In the present embodiment, when the second group is detected by the second detection unit 122, the selection unit 123 selects the second group regardless of whether or not the first group is detected by the first detection unit 121. select. When the second group is detected, it is assumed that a set of persons whose unknown relationship is established such as mutual communication is surrounding the robot apparatus 1. For example, since it is not appropriate to perform a service by a close conversation, the first group is not selected.

  When the second detection unit 122 does not detect the second group and the first detection unit 121 detects a plurality of first groups, the selection unit 123 sets the first group for each first group to be selected. Based on the movement vector belonging to the first group, an evaluation value indicating the suitability of the service target (ease of working if the viewpoint is changed) is calculated, and the first group having the highest evaluation value is selected. More specifically, it is as follows. First, the selection unit 123 extends a virtual straight line starting from the origin at every predetermined angle (for example, 5 degrees) on the XY plane with the robot apparatus 1 as the origin, and a plurality of first groups ( When the first group detected by the first detection unit 121 is present, the first group present at the position closest to the origin (the position closest to the robot apparatus 1) is selected as the selection target. In this way, the selection unit 123 selects the first group to be selected from the plurality of first groups detected by the first detection unit 121. When there is one first group selected as a selection target, the selection unit 123 selects the first group.

  On the other hand, when there are a plurality of first groups selected as selection targets, the selection unit 123 calculates an evaluation value for each first group to be selected based on a movement vector belonging to the first group. This will be specifically described below. For each first group to be selected, the selection unit 123 stores a history of detection data corresponding to a person included in the first group (a person corresponding to a movement vector belonging to the first group) as a person position storage unit 116. Read from. Then, for each person included in the first group, from the history of detection data corresponding to the person, the current time is reached from the past time point (for example, the time point 3 seconds before the current time) from the past by a certain time length. The movement vector recorded at each time point in the period until is acquired. Then, for each time point within the above period, an average value (average movement vector) of the movement vectors of each person at the time point is obtained and held as the movement vector transition Va of the first group.

  For each first group to be selected, the selection unit 123 calculates a reduction amount Da of the movement vector transition Va of the first group. More specifically, for example, for each time point within the above period (excluding the most past time point), the selection unit 123 calculates an average movement vector corresponding to the time point and an average movement vector corresponding to the time point immediately before the time point. And the average of each difference is held as the movement vector decrease amount Da.

  Next, the selection unit 123 weights the direction of the movement vector with the front direction of the robot apparatus 1 and the direction approaching the robot apparatus 1 as plus, and for each first group to be selected, the period For each time point, a vector direction evaluation value that increases as the direction of the average movement vector at that time point approaches the front direction of the robot apparatus 1 and the direction approaching the robot apparatus 1 is calculated. The average of the vector direction evaluation values calculated in step 1 is held as the movement vector direction evaluation value Db of the first group. In other words, the movement vector direction evaluation value Db of the first group shows a higher value as the direction of the movement vector is closer to the first direction toward the output unit (in this example, the speaker 161) that performs output related to the service. It can also be taken as.

Then, for each first group to be selected, the selection unit 123 calculates the calculated movement vector reduction amount Da, the movement vector direction evaluation value Db, and the number Num of persons corresponding to the movement vectors included in the first group. The number of faces detected by the face detection unit 111 and the relative distance Dis between the robot apparatus 1 and the first group are weighted and added, and the weighted addition result is used as an evaluation value (service for the first group). (Evaluation value indicating eligibility as a target to be conducted). The selection unit 123 selects the first group having the highest evaluation value from the first group to be selected. When the evaluation value is V, the weighted addition formula can be expressed by the following formula 2, for example.

  W1 to W5 in the above formula 2 are weighting factors, which may be constants or output values by functions. As understood from the above formula 2, the evaluation value V of the first group increases as the movement vector decrease amount Da of the first group increases. The movement vector decrease amount Da of the first group is large because it is estimated that there is a high possibility that the group of persons belonging to the first group is interested in the robot apparatus 1 and is standing. . Moreover, the evaluation value V of the said 1st group becomes large, so that the movement vector direction evaluation value Db of the 1st group is large. The fact that the movement vector direction evaluation value Db of the first group is large is because it is estimated that there is a high possibility that the robot apparatus 1 is approaching with interest. In order to increase the effect of providing the service, it is preferable to provide the service to a set of persons who are highly likely to be interested in the robot apparatus 1, so the movement vector reduction amount Da and the movement vector direction evaluation value The first group having a large value of Db is considered to have high eligibility as a service target, and the evaluation value V becomes high. Further, as can be understood from Equation 2, the evaluation value V indicates a higher value as the number of detected faces increases. Since the number of detected faces is large, it is presumed that there are many persons who are turning their faces toward the robot apparatus 1, that is, there is a high possibility that they are interested in the robot apparatus 1. This is because it is considered to be highly qualified as a subject to be conducted.

  When the first group is selected as described above, the selection unit 123 determines that the distance between the selected first group and the robot apparatus 1 exceeds the distance that allows interaction (for example, 3.6 m or less). Stop the selection process. In the present embodiment, when the distance between the first group having the highest evaluation value V and the robot apparatus 1 exceeds the distance at which dialogue is possible, the selection unit 123 selects the first group having the highest evaluation value V. Without doing so, the above selection process is performed again. Further, when the distance between the selected first group and the robot apparatus 1 exceeds the distance at which the conversation can be performed, the selection unit 123 temporarily cancels the selection of the first group, and restarts the above selection process. Re-select a new first group. Moreover, the selection part 123 can cancel | release the selection of the 1st group, and can restart the above-mentioned selection process, when the evaluation value V of the selected 1st group becomes below a 4th threshold value.

  When selecting the first group, the selection unit 123 notifies the initial operation control unit 130 and the first determination unit 140 of a first label number that can identify the selected first group. When the second group is selected, the selection unit 123 notifies the initial operation control unit 130 and the first determination unit 140 of a second label number that can identify the selected second group.

  Returning to FIG. 3, the description will be continued. When the selection unit 123 selects the first group, the initial operation control unit 130 controls the execution of the initial operation (operation before performing the service) for the selected first group. More specifically, it is as follows.

  The initial operation control unit 130 includes a second determination unit 132, a voice call content storage unit 133, a group tracking unit 134, and a motor control unit 135. When the selection unit 123 selects the first group, the second determination unit 132 determines the content of the initial operation performed before performing the service. The second determination unit 132 uses the first label number notified from the selection unit 123, and is included in the first group (the first group selected by the selection unit 123) identified by the first label number. Is detected from the person position storage unit 116. When there is a person associated with the attribute information among the persons included in the selected first group, the second determination unit 132 refers to the call content storage unit 133 and determines the voice according to the attribute information. Determine (select) the details of the call. That is, the second determination unit 132 determines the content of the initial operation to be performed before performing the first service, according to the attribute determined by the attribute determination unit 112.

  In the present embodiment, as shown in FIG. 6, the voice call content storage unit 133 stores a plurality of voice call data (voice data in this example) for each combination of gender and age. In FIG. 6, “F1-2” represents a combination of the sex “female” and the age “10 to 20”. “F3-4” represents a combination of the sex “female” and the age “30s to 40s”. “F5-6” represents a combination of the sex “female” and the age “50s-60s”. “M1-2” represents a combination of a sex “male” and an age “10's to 20's”. “M3-4” represents a combination of the sex “male” and the age “30s to 40s”. “M5-6” represents a combination of the sex “male” and the age “50s-60s”. For example, when the attribute information is a combination of the gender “female” and the age “10's to 20's”, the second determination unit 132 includes a plurality of voice call data associated with “F1-2”. The voice call data is determined as the voice call content. In the example of FIG. 6, the second determining unit 132, "Hello, Miss" voice over data that can also be determined as a voice over contents.

  Further, the second determination unit 132 outputs the direction related to the selected movement vector belonging to the first group, the front direction of the robot apparatus 1 and the direction approaching the robot apparatus 1 (if the view is changed, the service-related output is changed). The content of the voice call can be determined according to the degree of coincidence with the first direction toward the output unit to be performed. For example, when the direction of the movement vector crosses the front of the robot apparatus 1 and the degree of coincidence is low, the voice call data that calls the moving person may be determined as the voice call content. it can. On the other hand, when the direction of the movement vector is the direction approaching the robot apparatus 1 and the above-mentioned degree of coincidence is high, the voice call data on the assumption that the person is paying attention to this is used as the voice call content. It can also be determined.

  The voice call data determined (selected) by the second determination unit 132 as described above is output from the speaker 161. The second determination unit 132 can arbitrarily set the timing for outputting the determined voice call data. For example, the second determination unit 132 outputs the determined voice call data to the speaker at the timing when the selected first group evaluation value V becomes equal to or greater than a certain value. It is also possible to perform control to output from 161. In the present embodiment, the above initial operation (voice call operation) is performed only once for one selected first group.

  The group follower 134 calculates the distance between the position coordinate Pa of the selected first group and the robot apparatus 1, and if the distance is less than or equal to a distance that allows dialogue (for example, 3.6 m or less), the first camera 101. A command is output to the motor control unit 135 so that the image capturing direction is directed to the position coordinate Pa. The motor control unit 135 controls driving of the driving unit 170 in accordance with a command from the group tracking unit 134. Thereby, the line of sight from the first camera 101 can be directed to the selected first group.

  Returning to FIG. 3, the description will be continued. When the selection unit 123 selects the first group, the first determination unit 140 determines to perform the first service, and when the selection unit 123 selects the second group, what is the first service? Decide to perform a different second service. In the present embodiment, the first service is to provide interactive information, and the second service is to provide non-interactive information. This will be specifically described below. In addition, when the first determination unit 140 receives a notification from the selection unit 123 to cancel the selection of the first group, the first determination unit 140 determines not to perform the service for the first group.

  When the selection unit 123 selects the first group, the first determination unit 140 uses the first label number notified from the selection unit 123 to identify the first group identified by the first label number (selection unit The detection data of the persons included in the first group selected by 123 is acquired from the person position storage unit 116. Then, the first determination unit 140 refers to the application-specific application storage unit 141 and determines (selects) the content of the first service according to the number of persons included in the selected first group. Further, when the second group is selected by the selection unit 123, the first determination unit 140 refers to the application storage unit 141 for each use and determines (selects) the content of the second service.

  In this embodiment, as shown in FIG. 7, the application-specific application storage unit 141 is activated when the selected group is the first group and the number of persons included in the first group is less than three. A plurality of applications associated with an individually corresponding application (hereinafter referred to as “individually-applicable APP”) and a selected group are the first group, and the number of persons included in the first group is Multiple applications associated with a group-compatible application (hereinafter referred to as “group-compatible APP”) indicating an application to be activated when there are three or more persons, and activated when the selected group is the second group A plurality of applications associated with a multi-person application (hereinafter referred to as “multi-person APP”) indicating It is.

  When the group selected by the selection unit 123 is the first group and the number of persons included in the first group is less than three, the first determination unit 140 includes a plurality of applications associated with the individual corresponding APP. Is selected. In addition, the first determination unit 140 includes a plurality of groups associated with the group-corresponding APP when the group selected by the selection unit 123 is the first group and the number of persons included in the first group is three or more. One of the applications is selected. Furthermore, when the group selected by the selection unit 123 is the second group, the first determination unit 140 selects any one of a plurality of applications associated with the multi-person APP. The application selected by the first determination unit 140 is executed by the application execution unit 142.

The individual correspondence APP is an application for performing a first service that provides interactive information. When any of the plurality of applications associated with the individual corresponding APP is selected by the first determination unit 140, the application execution unit 142 executes the selected application. Thereby, the speaker 161 and the drive part 170 perform the output according to the selected application, and a 1st service is implement | achieved. As an example of the first service in this case, information is exchanged (such as product information or facility guidance information) after a close exchange such as asking a question multiple times and accepting an answer to the question. Can be provided while exchanging the behavior of the robot apparatus 1.
In the present embodiment, the first determination unit 140 associates with the individual correspondence APP when the group selected by the selection unit 123 is the first group and the number of persons included in the first group is less than three. By selecting one of the plurality of applications, it is determined to perform the first service as described above.

  In addition, the group-compatible APP is an application for realizing the first service that provides interactive information, as with the individual-compatible APP. However, since the number of people to whom the service is provided increases, As an example of the first service realized by execution, an average request is obtained by using a question or attribute information whose answer is a majority decision, and information corresponding to the obtained request is obtained from the robot apparatus 1. It is possible to provide it while providing the behavior of. In the present embodiment, when the group selected by the selection unit 123 is the first group and the number of persons included in the first group is three or more, the first determination unit 140 associates with the group correspondence APP. By selecting one of the plurality of applications, it is determined to perform the first service as described above.

  In addition, when executing individually-applicable APP, dialogue using voice recognition is effective. However, when executing group-applicable APP, it is predicted that the utterances of each person collide and recognition becomes difficult. Therefore, it is effective to use face attribute recognition or gesture recognition using an image. In addition, regarding the way of gazing of the robot apparatus 1 that is executing the application, when executing the individually corresponding APP, direct the line of sight to a specific person, and when executing the group corresponding APP, For each, the way of gazing may be changed depending on the number of people, such as switching the sight line sequentially.

  The multi-person APP is an application for performing a second service that provides non-interactive information. When the first determination unit 140 selects any one of a plurality of applications associated with the multi-person APP, the application execution unit 142 executes the selected application. Thereby, the speaker 161 and the drive part 170 perform the output according to the selected application, and a 2nd service is implement | achieved. As an example of the second service in this case, without performing a dialogue or the like (regardless of the request from the side receiving the service provision), information that advertises a product or a store is used for the dramatic behavior of the robot apparatus 1 ( It is possible to provide the information while providing a predetermined operation). In this case, the way of swinging the line of sight of the robot apparatus 1 may be in a direction directed to the advertising effect rather than following the person. In the present embodiment, when the group selected by the selection unit 123 is the second group, the first determination unit 140 selects one of a plurality of applications associated with the multi-person APP. Then, it is determined to perform the second service as described above.

  In the present embodiment, the application-specific application storage unit 141 is connected to the store terminal 200 via the communication unit 185, and the application stored in the application-specific application storage unit 141 has contents according to the application at the store. It can also be changed. In addition, it is not necessary to change the content of the application memorize | stored in the application storage part 141 classified by use, and the form in which the shop terminal 200 and the communication part 185 are not provided may be sufficient.

  Here, the service that provides interactive information (first service) refers to a service that provides information while interactively exchanging between the service providing side and the service receiving side. . “Interaction” here refers to the mutual transmission of linguistic information and non-linguistic information (for example, gestures) of people or robots. Acquisition and output of information that does not change in the short term is “ It is not included in “Interaction”. For example, when acquiring information such as age and gender determined from the face of the person to whom the service is provided and providing product information according to the acquired information, it is determined that no "exchange" has occurred. The On the other hand, when acquiring information that has changed in the short term, such as changes in facial expression of the person to whom the service is provided, and providing product information according to the acquired information, To be judged. Further, the service that provides non-interactive information (second service) refers to a service that provides information without exchanging between the service providing side and the service receiving side.

  Returning to FIG. 3, the description will be continued. The output unit 150 performs output according to the determination by the first determination unit 140 or the second determination unit 132. The output unit 150 includes a speaker 161 and a drive unit 170. The speaker 161 performs audio output according to the application determined by the first determination unit 140 or the voice call data determined by the second determination unit 132. The driving unit 170 performs an operation output according to the application determined by the first determination unit 140. Further, the drive unit 170 performs an operation output according to a command from the motor control unit 135.

  The face detection unit 111, the attribute determination unit 112, the person detection unit 113, the person position calculation unit 114, the movement vector detection unit 115, the acquisition unit 117, the first detection unit 121, the second detection unit 122, and the selection unit 123 described above. Each of the functions of the second determination unit 132, the group tracking unit 134, the motor control unit 135, the first determination unit 140, and the application execution unit 142 is executed by the CPU 181 in FIG. 2 by executing a program stored in the memory 182 or the HDD 183. It is realized by executing. For example, the face detection unit 111, the attribute determination unit 112, the person detection unit 113, the person position calculation unit 114, the movement vector detection unit 115, the acquisition unit 117, the first detection unit 121, and the second detection described above. Configuration in which at least a part of unit 122, selection unit 123, second determination unit 132, group tracking unit 134, motor control unit 135, first determination unit 140, and application execution unit 142 is realized by a dedicated hardware circuit It may be.

  FIG. 8 is a flowchart illustrating an example of a processing operation performed by the robot apparatus 1 according to the present embodiment. In step S100, the group detection / selection unit 120 detects the first group or the second group, and selects either the detected first group or the second group. When the selection unit 123 selects either the first group or the second group (result of step S101: YES), the process proceeds to step S102. When the group selected by the selection unit 123 is the second group (result of Step S102: NO), the first determination unit 140 determines to perform the second service that provides non-interactive information, From the application-specific application storage unit 141, one of a plurality of applications associated with the multi-person APP is selected. Then, the application execution unit 142 executes the selected application (step S103).

  On the other hand, in the above-described step S102, when the group selected by the selection unit 123 is the first group (result of step S102: YES), the initial operation control unit 130 selects the first group selected by the selection unit 123. The execution of the initial operation is controlled (step S104). Next, the first determination unit 140 determines whether the number of persons included in the first group selected by the selection unit 123 is less than three (step S105). When it is determined that the number of persons included in the first group is less than three (result of step S105: YES), the first determination unit 140 corresponds to the individual correspondence APP from the application storage unit 141 by use. One of a plurality of attached applications is selected. Then, the application execution unit 142 executes the selected application (step S106).

  When it is determined in step S105 described above that the number of persons included in the first group is three or more (result of step S105: NO), the first determination unit 140 uses the application storage unit by usage. From 141, one of a plurality of applications associated with the group-compatible APP is selected. Then, the application execution unit 142 executes the selected application (step S107).

  In the present embodiment, when the first group is selected by the selection unit 123, the initial operation is performed on the selected first group. However, the present invention is not limited thereto, and for example, the initial operation is not performed. Form may be sufficient.

  As described above, in the present embodiment, the first group that provides interactive information when the first group that has a size equal to or larger than the threshold and indicates a set of similar movement vectors is selected. While deciding to perform a service, if a second group representing a set of motion vectors whose magnitude is less than a threshold is selected, it is determined to perform a second service that provides non-interactive information. In other words, when the service provision target is a set of persons who have established a relationship such as communicating with each other, it is determined to perform the first service that provides information based on close communication, If the target of the service is a set of people whose relationship is unknown, such as communicating with each other, a service that provides information unilaterally (non-interactive nature) Therefore, an appropriate service can be performed according to the situation of the service provision target.

(Second Embodiment)
In the above-described first embodiment, a form in which a service is performed using a robot is exemplified. However, in this embodiment, a display in which characters represented by CG (computer graphics) are displayed on the screen instead of the robot. Use to service. In the present embodiment, the above-described driving unit 170 is replaced with a display, and the behavior of the robot by the driving unit 170 is expressed by CG animation or the like. The service providing apparatus according to the present embodiment has a configuration in which the driving unit 170 is replaced with a display in the configuration of the robot apparatus 1 described above.

  The CG character displayed on the screen may have at least a symbol indicating an eye and can change its eye direction. In addition, the language information may be output by voice output with a speaker attached to the display, or may be displayed as a character by adding a speech balloon to the CG character displayed on the display. Further, when detecting a group, detection is performed at a fixed distance from the own apparatus. However, in the case of a flat surface such as a display, the groups to be detected are arranged in a straight line. However, since an advertisement in an elevator or the like may be enclosed on an arc, the center point or plane of the display is selected as a reference for the distance from the device itself. Further, the fixed distance may be changed depending on the size of the display or the height of installation.

  The control of the line of sight of the CG character is performed by accumulating animation data in advance, calculating a relative angle from the person position calculated by the person position calculating unit 114, and calling and displaying an animation facing the angle. . If the animation data is a CG character that can be generated sequentially (for example, a character represented by a polygon having three-dimensional information), the data may be displayed while being generated. In the scene where information is finally provided, by displaying the product / facility information and the CG character at the same time, it is possible to obtain an advertisement effect with high visual transmission. In addition, as an information input device from a person, a touch panel or a haptic device may be installed on a display to perform tactile communication. With the above configuration, the same behavior as that of the robot can be performed, and the same contents as those of the first embodiment can be realized by a display.

  As mentioned above, although embodiment of this invention was described, each above-mentioned embodiment was shown as an example and is not intending limiting the range of invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

  For example, the functions of the above-described person movement vector detection unit 110 (the face detection unit 111, the attribute determination unit 112, the person detection unit 113, the person position calculation unit 114, the movement vector detection unit 115, and the person position storage unit 116) are the functions of the robot apparatus 1. It may be a form extracted from and mounted on an external device (server device). In this case, the acquisition unit 117 acquires a movement vector of a person existing around the robot apparatus 1 (existing in the camera detection range) from the external apparatus. In short, the information processing apparatus according to the present invention includes an acquisition unit 117 that acquires movement vectors of each of a plurality of objects, and a first group that indicates a set of movement vectors that are similar in size to each other and that have a size equal to or larger than a threshold value. A first detection unit 121 for detecting the second detection unit 122 for detecting a second group indicating a set of movement vectors whose magnitude is less than a threshold, and a selection for selecting either the first group or the second group When the first group is selected by the selection unit 123 and the selection unit 123, it is determined to perform the first service, and when the second group is selected by the selection unit 123, the second service is determined to be performed. The 1st determination part 140 to be performed should just be a form provided. In addition to the configuration of the information processing apparatus described above, the service providing apparatus according to the present invention only needs to further include an output unit 150 that performs output according to the determination by the first determination unit 140.

  Further, for example, the function of the information processing apparatus 100 described above may be distributed to the information processing apparatus and one or more server apparatuses. That is, the information processing apparatus 100 described above can be replaced with an information processing system in which the information processing apparatus and one or more server apparatuses communicate with each other. For example, the face detection unit 111, the attribute determination unit 112, the person detection unit 113, the person position calculation unit 114, the movement vector detection unit 115, the person position storage unit 116, the acquisition unit 117, the first detection unit 121, and the second detection unit described above. 122, the selection unit 123, the second determination unit 132, the calling content storage unit 133, the group tracking unit 134, the motor control unit 135, the first determination unit 140, the application-specific application storage unit 141, and the application execution unit 142. However, the information processing apparatus and one or more server apparatuses may be distributed and mounted. In short, the present invention is an information processing system including an information processing device and one or more server devices, and an acquisition unit 117 that acquires a movement vector of each of a plurality of objects, and a size that is equal to or larger than a threshold value. In addition, a first detection unit 121 that detects a first group indicating a set of movement vectors similar to each other, and a second detection unit 122 that detects a second group indicating a set of movement vectors whose magnitude is less than a threshold value. When the first unit is selected by the selection unit 123 and the selection unit 123 that selects either the first group or the second group, it is determined to perform the first service, and the selection unit 123 performs the second operation. When the group is selected, the first determination unit 140 that determines to perform the second service may be used.

  Further, for example, each function of the above-described information processing apparatus 100 is distributed and installed in one or more server apparatuses. Among the configurations of the one or more server apparatuses and the robot apparatus 1 described above, the information processing apparatus 100 is included. The form which the apparatus (output device) comprised from the element excepted may mutually communicate may be sufficient. That is, the above-described robot device 1 can be replaced with a service providing system including one or more server devices and an output device. In short, the present invention is a service providing system including one or more server devices and an output device, and the one or more server devices include an acquisition unit 117 that acquires each movement vector of a plurality of objects, and a large size. A first detection unit 121 for detecting a first group indicating a set of similar movement vectors whose length is equal to or greater than a threshold, and a second group indicating a set of movement vectors having a size less than the threshold The second detection unit 122, the selection unit 123 that selects either the first group or the second group, and the first unit is selected by the selection unit 123, the first service is determined, When the second group is selected by the selection unit 123, a first determination unit 140 that determines to perform the second service is provided, and the output device performs output according to the determination by the first determination unit 140 Less output unit 150 It may be any form provided also.

  Note that the program executed by the CPU 181 may be provided by storing it on a computer connected to a network such as the Internet and downloading it via the network. The program executed by the CPU 181 may be provided or distributed via a network such as the Internet. The program executed by the CPU 181 may be provided by being incorporated in advance in a nonvolatile recording medium such as a ROM.

DESCRIPTION OF SYMBOLS 1 Robot apparatus 10 Head 20 Right arm part 30 Left arm part 40 Body part 50 Right leg part 60 Left leg part 100 Information processing apparatus 101 1st camera 102 2nd camera 110 Person movement vector detection part 111 Face detection part 112 Attribute determination part 113 Human detection unit 114 Human position calculation unit 115 Movement vector detection unit 116 Human position storage unit 117 Acquisition unit 120 Group detection / selection unit 121 First detection unit 122 Second detection unit 123 Selection unit 130 Initial operation control unit 132 Second determination unit 133 Calling content storage unit 134 Group follow-up unit 135 Motor control unit 140 First determination unit 141 Application-specific application storage unit 142 Application execution unit 150 Output unit 161 Speaker 162 Microphone 170 Drive unit 171T Motor 172P Motor 177W Motor 181 CPU
182 Memory 183 HDD
184 External I / O
185 Communication unit 200 Store terminal

The information processing apparatus according to the embodiment includes an acquisition unit, a first detection unit, a second detection unit, a selection unit, and a first determination unit. The acquisition unit acquires each movement vector of the plurality of objects. Based on the similarity between the movement vector of the object and the movement vector of another object existing within the range defined by the personal space of the object , the first detection unit has a similarity that is greater than or equal to a reference value . A first group indicating a set of movement vectors similar to each other is detected. The second detection unit detects a second group that represents a set of movement vectors that exist within a predetermined range from the service providing apparatus and that have a size less than a threshold value. The selection unit selects either the first group or the second group. The first determination unit, if the first group is selected by the selection unit, decides to perform the first service that provides information interactively when the second group is selected by the selection unit, the non Decide to perform a second service that provides interactive information .

An information processing system according to an embodiment is an information processing apparatus that controls an operation of a service providing device , and one or more server devices , and an acquisition unit, a first detection unit, and a second detection unit are selected. And a first determination unit. The acquisition unit acquires each movement vector of the plurality of objects. Based on the similarity between the movement vector of the object and the movement vector of another object existing within the range defined by the personal space of the object , the first detection unit has a similarity that is greater than or equal to a reference value . A first group indicating a set of movement vectors similar to each other is detected. The second detection unit detects a second group that represents a set of movement vectors that exist within a predetermined range from the service providing apparatus and that have a size less than a threshold value. The selection unit selects either the first group or the second group. The first determination unit, if the first group is selected by the selection unit, decides to perform the first service that provides information interactively when the second group is selected by the selection unit, the non Decide to perform a second service that provides interactive information .

The service providing system according to the embodiment is a service providing system including one or more server devices and an output device, and the one or more server devices include an acquisition unit, a first detection unit, a second detection unit, and a selection unit. A determination unit. The acquisition unit acquires each movement vector of the plurality of objects. Based on the similarity between the movement vector of the object and the movement vector of another object existing within the range defined by the personal space of the object, the first detection unit has a similarity that is greater than or equal to a reference value. A first group indicating a set of movement vectors similar to each other is detected. The second detection unit detects a second group that represents a set of movement vectors that exist within a predetermined range from the output device and that have a magnitude less than a threshold value. The selection unit selects either the first group or the second group. The determining unit determines to perform the first service that provides interactive information when the first group is selected by the selecting unit, and is non-interactive when the second group is selected by the selecting unit. The second service providing the information is determined to be performed. On the other hand, the output device includes at least an output unit that performs output according to the determination by the determination unit.

The information processing method of the embodiment includes an acquisition step, a first detection step, a second detection step, a selection step, and a first determination step. In the acquisition step, each movement vector of the plurality of objects is acquired. In the first detection step , based on the similarity between the movement vector of the object and the movement vector of another object existing within the range defined by the personal space of the object, the similarity is equal to or greater than a reference value . A first group indicating a set of movement vectors similar to each other is detected. The second detection step detects a second group indicating a set of movement vectors that are within a predetermined range from the service providing apparatus and that have a magnitude less than a threshold value. The selection step selects either the first group or the second group. The first determining step determines to perform a first service that provides interactive information when the first group is selected, and provides non-interactive information when the second group is selected. It determines the execution of the second service.

Claims (15)

An acquisition unit for acquiring a movement vector of each of a plurality of objects;
A first detection unit for detecting a first group having a set of movement vectors similar in size to each other, the magnitude of which is equal to or greater than a threshold;
A second detector for detecting a second group indicating the set of movement vectors that is less than the threshold;
A selection unit for selecting either the first group or the second group;
When the first group is selected by the selection unit, the first service is determined to be performed, and when the second group is selected by the selection unit, a second service different from the first service is determined. A first determination unit that determines to perform
Information processing device. The first service is to provide interactive information;
The information processing apparatus according to claim 1. The second service is to provide non-interactive information;
The information processing apparatus according to claim 1. The object is a person;
A face detection unit for detecting a face of the object from an image captured by the imaging device;
The second detection unit has a number of objects whose faces are detected by the face detection unit among a plurality of the objects corresponding one-to-one with the plurality of movement vectors having a size less than the threshold. A plurality of movement vectors whose magnitude is less than the threshold value are detected as the second group in the case of two or more threshold values;
The information processing apparatus according to claim 1. When the number of the movement vectors belonging to the set of movement vectors similar to each other is equal to or larger than a third threshold, the first detection unit determines the set as the first group. Not detect as,
The information processing apparatus according to claim 1. When the second detection unit detects the second group, the selection unit selects the second group regardless of whether or not the first detection unit detects the first group. ,
The information processing apparatus according to claim 1. If the second detection unit detects a plurality of the first groups without the second detection unit being detected by the second detection unit, the selection unit selects each of the first groups to be selected. , Based on the movement vector, to calculate an evaluation value indicating eligibility as a target to be serviced, and select the first group having the highest evaluation value,
The information processing apparatus according to claim 1. The evaluation value indicates a higher value as the direction of the movement vector is closer to the first direction toward the output unit that performs output according to the determination by the first determination unit.
The information processing apparatus according to claim 7. The object is a person;
A face detection unit for detecting a face of the object from an image captured by the imaging device;
The evaluation value indicates a higher value as the number of faces detected by the face detection unit is larger.
The information processing apparatus according to claim 7. When the first group is selected by the selection unit, the direction toward the movement vector included in the selected first group and the output unit that performs output according to the determination by the first determination unit A second determination unit configured to determine the content of an initial operation performed before performing the first service according to the degree of coincidence with one direction;
The information processing apparatus according to claim 1. The object is a person;
A face detection unit that detects a face of the object from an image captured by an imaging device;
An attribute determination unit that determines an attribute of the object whose face is detected by the face detection unit,
When the first group is selected by the selection unit, a second determination unit that determines the content of the initial operation performed before performing the first service according to the attribute determined by the attribute determination unit Prepare
The information processing apparatus according to claim 1. An output unit that performs output according to the determination by the first determination unit;
The information processing apparatus according to claim 1. An information processing system comprising an information processing device and one or more server devices,
An acquisition unit for acquiring a movement vector of each of a plurality of objects;
A first detection unit for detecting a first group having a set of movement vectors similar in size to each other, the magnitude of which is equal to or greater than a threshold;
A second detector for detecting a second group indicating the set of movement vectors that is less than the threshold;
A selection unit for selecting either the first group or the second group;
When the first group is selected by the selection unit, the first service is determined to be performed, and when the second group is selected by the selection unit, a second service different from the first service is determined. A determination unit that determines to perform
Information processing system. A service providing system comprising one or more server devices and an output device,
The one or more server devices are:
An acquisition unit for acquiring a movement vector of each of a plurality of objects;
A first detection unit for detecting a first group having a set of movement vectors similar in size to each other, the magnitude of which is equal to or greater than a threshold;
A second detection unit for detecting a second group indicating the set of movement vectors that is less than a threshold;
A selection unit for selecting either the first group or the second group;
When the first group is selected by the selection unit, the first service is determined to be performed, and when the second group is selected by the selection unit, a second service different from the first service is determined. A decision unit that decides to perform
The output device is
At least an output unit that performs output according to the determination by the determination unit;
Service provision system. An obtaining step for obtaining a movement vector of each of the plurality of objects;
A first detection step of detecting a first group having a set of movement vectors similar in size to each other, the magnitude of which is equal to or greater than a threshold;
A second detection step of detecting a second group indicating the set of movement vectors that is less than the threshold;
A selection step of selecting either the first group or the second group;
When the first group is selected, it is determined to perform the first service, and when the second group is selected, the first service is determined to perform the second service different from the first service. A determination step,
Information processing method.

Images