JP2005275935A - Terminal device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To raise the convenience of a user interface in a device which is usable even in a place with much noise such as a busy place without requiring a card which records information concerning users. <P>SOLUTION: The facial image of the user is picked up so as to estimate the attribute (race, sex, age, etc.) of the user through the use of the image. The convenience of the user interface is raised by controlling the content or state of an output, on the basis of the estimation result. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a technique that is effective when applied to a device that outputs information, options, and the like to a user.

  Conventionally, the language used for display and voice output in a terminal device has been determined in advance according to the environment in which the terminal device is installed. However, as the globalization progresses, the language is not limited to the installation environment. It has become necessary to support multiple languages. For example, in an ATM (Automated Teller Machine) in the United States, since it corresponds to many languages, it is possible to change the subsequent display to the display of the language selected by the user by operating the touch panel and buttons. Yes. However, there is a problem that as the number of language choices increases, a user needs more time and effort to select a desired language.

For such a problem, there is a technique for acquiring user information recorded in an ID card, a credit card, a bankbook, etc. as a person identification information input means and switching a language to be used according to the characteristics of the user information. Yes (see Patent Document 1). Further, there is a technique for recognizing the user's first voice, identifying the user's language, and responding by voice response using the user's language (see Patent Document 2). According to these technologies, it is not necessary for the user to bother to select a desired language, and it is possible to reduce the labor required for operating the terminal device.
JP 2000-250840 A JP 2003-316383 A

  However, the conventional technique has the following problems. In the technique for recording user information, the user needs to have a recording medium such as a card corresponding to the terminal device to be used. For this reason, user information is not provided to terminal devices that do not require identification of users (for example, public terminal devices, such as ticket vending machines, tourist information devices, and explanation devices such as museums and art galleries). It was difficult to apply the technology to record.

  Also, with the technology for recognizing the language from the user's voice, it is difficult to accurately recognize the user's voice in a noisy place such as a busy place. In particular, the above-mentioned device that does not need to specify a user is often installed in such a noisy place, and thus a technique for recognizing a language from a user's voice has been inappropriate.

  Therefore, the present invention solves these problems and is an apparatus that can be used in a noisy place such as a place where there is a lot of traffic without requiring a card or the like on which information about the user is recorded. An object of the present invention is to provide an apparatus for improving performance.

  In order to solve the above problems, the present invention has the following configuration. 1st aspect of this invention is a terminal device, Comprising: The imaging means which images a user's face, The detection means which detects a user's face from the image imaged by the imaging means, The image of the face detected by the detection means Attribute estimation means for estimating the user's attributes, output control means for controlling the content and mode of output according to the estimation result by the attribute estimation means, and output means for outputting according to the control content by the output control means .

  According to the present invention, the attribute is estimated from the image of the user's face without using the user's voice or information recorded on the card. Therefore, the user does not need to hold a recording medium such as a card, and the user's attribute can be estimated even in a noisy place. And according to this invention, the content and aspect of an output are controlled according to the estimated user attribute (estimation result).

  Here, the user attribute is information on the race, age, sex, etc. of the user, for example. The contents and form of output include, for example, the language and words used for output, the contents of output text, the size of displayed characters, the presence or absence of phonetics for displayed characters, and the displayed image. The content and type, the volume of the output sound, and the gender of the output sound. Since the content and mode of such output are controlled according to the user's attributes, it is possible to improve the convenience of the user interface of the terminal device.

  Moreover, the attribute estimated by the attribute estimation means may be configured to include information on the race of the user. In this case, the output control means may be configured to control the output means so as to perform output in a language corresponding to the estimated race.

  In the present invention configured as described above, at least information on the race of the user is estimated. And since the output by the language according to the estimated race of the user is performed, the user can easily understand the contents of the output.

  Further, when the attribute estimated by the attribute estimation unit is configured to include information on the user's race, the output control unit outputs so as to output a plurality of language candidates according to the estimated race. It may be configured to control the means. In this case, the terminal device may further include a selection unit for the user to select a language used in the output unit from a plurality of language candidates. Further, the output control means may be configured to control the output means so as to perform output in the selected language after the language is selected via the selection means.

  In the present invention configured as described above, the user can select a language used in the output unit by using the selection unit. At this time, not all languages are presented as candidates for selection, but a plurality of languages according to the race of the user are presented as candidates. Language candidates corresponding to the race may be determined in advance based on, for example, a statistical result of a language used by each race. With this configuration, it is possible to reduce the labor when the user selects the language to be used.

  The attribute may be configured to include information related to the user's age. In this case, the output control means displays the displayed characters larger than normal when the estimated age can be determined to be older, and / or the output sound is louder than normal. It may be configured to control the output means so as to output the output.

  In the present invention configured as described above, the age of the user is estimated, and when it can be determined that the user is old from this age, the characters are displayed larger than the normal time. Or when it can be judged that a user is elderly, a sound is output with a louder volume than usual. For this reason, the output is performed at a volume that is easy to read and easy to hear for the elderly.

In addition, when the estimation processing for a plurality of faces is performed by the attribute estimation unit, the output control unit further includes an estimation result selection unit that selects which of the plurality of types of estimation results should be used for processing. You may comprise so that it may provide. In this case, the output control means may be configured to perform processing using the estimation result selected by the estimation result selection means.

  The present invention may be realized by a program being executed by an information processing apparatus. That is, the present invention can specify the processing executed by each of the above-described means as a program for causing the information processing apparatus to execute, or a recording medium on which the program is recorded. Further, the present invention may be specified by a method in which the information processing apparatus executes the processing executed by each of the above-described means.

  According to the present invention, since the attribute is estimated using the face image of the user, it is not necessary to use a card or the like on which information about the user is recorded, and even in a noisy place such as a busy place, It is possible to estimate the user's attributes and improve the convenience of the user interface according to the attributes.

  In the following description, a face image is an image including at least a part or all of an image of a person's face. Therefore, the face image may include an image of the entire person, or may include an image of only the face of the person or only the upper body. Moreover, you may include the image about a several person. Furthermore, the background may include any pattern such as a scenery (background: including objects that have attracted attention as a subject) or a pattern other than a person.

[System configuration]
In terms of hardware, the terminal device 1 includes a CPU (Central Processing Unit), a main storage device (RAM), an auxiliary storage device, and the like connected via a bus. The auxiliary storage device is configured using a nonvolatile storage device. The non-volatile storage device referred to here is a so-called ROM (including Read-Only Memory: EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), mask ROM, etc.),
FRAM (Ferroelectric RAM), hard disk, etc.

  FIG. 1 is a diagram illustrating functional blocks of the terminal device 1. The terminal device 1 loads various programs (OS, applications, etc.) stored in the auxiliary storage device into the main storage device and is executed by the CPU, whereby the imaging unit 2, the face detection unit 3, the feature extraction unit 4, It functions as a device including an attribute estimation unit 5, an attribute information determination unit 6, an output control unit 7, an output unit 8, and the like. The face detection unit 3, the feature extraction unit 4, the attribute estimation unit 5, the attribute information determination unit 6, and the output control unit 7 are realized by executing a program by the CPU. Further, the face detection unit 3, the feature extraction unit 4, the attribute estimation unit 5, the attribute information determination unit 6, and the output control unit 7 may be configured as a dedicated chip. Next, each functional unit included in the terminal device 1 will be described.

(Imaging section)
The imaging unit 2 is configured to capture a face image of the user of the terminal device 1, and passes the captured image to the face detection unit 3. For example, the imaging unit 2 is configured using a CCD (Charge-Coupled Devices), a CMOS (Complementary Metal-Oxide Semiconductor) sensor, or the like.

  The imaging unit 2 may be configured to always perform imaging, or may be configured to perform imaging when it is detected by a sensor (not illustrated) that the user intends to use the terminal device 1. However, it may be configured to perform imaging when the terminal device 1 is operated by the user.

(Face detection unit)
The face detection unit 3 detects a human face from the image captured by the imaging unit 2 and specifies face information indicating the position and size of the detected face. For example, the face detection unit 3 may be configured to detect a face by template matching using a reference template corresponding to the contour of the entire face. The face detection unit 3 may be configured to detect a face by template matching based on face components (eg, eyes, nose, ears). The face detection unit 3 may be configured to detect a vertex such as a head by chroma key processing and detect a face based on the vertex. The face detection unit 3 may be configured to detect a region close to the skin color and detect the region as a face. Further, the face detection unit 3 may be configured to perform learning by a teacher signal using a neural network and detect a face-like region as a face. In addition, the face detection process by the face detection unit 3 may be realized by applying any existing technique.

  Further, when a plurality of human faces are detected from the image captured by the imaging unit 2, the face detection unit 3 may determine a face to be processed according to a specific criterion. The predetermined reference is, for example, the size of the face, the direction of the face, the position of the face in the image, a distance measuring unit (not shown) (for example, distance measurement is performed by emitting a light beam to an object and obtaining reflected light thereof). The distance measurement result by means).

  The face detection unit 3 passes face information regarding the detected face to the feature extraction unit 4.

(Feature extraction unit)
First, the feature extraction unit 4 sets a plurality of feature points in the face detected by the face detection unit 3 (feature point setting processing). Then, the feature extraction unit 4 acquires the feature quantity of each feature point as the feature quantity of the face of the subject based on the feature point set by the feature point setting process (feature quantity acquisition process). Hereinafter, feature point setting processing and feature amount acquisition processing will be described.

<Feature point setting process>
In the feature point setting process, the feature extraction unit 4 first detects the detected facial organs. Examples of facial organs include eyes, nose, nostrils, mouth (lips), eyebrows, jaws, and forehead. The feature extraction unit 4 may detect any facial organ or a plurality of organs. For example, the feature extraction unit 4 is configured to detect both eyes and mouth of the face of the subject.

  Next, the feature extraction unit 4 converts the detected face image into a grayscale image. The feature extraction unit 4 also performs angle normalization and size normalization of the detected face image based on the positional relationship of the detected facial organs. These processes are collectively called preprocessing. Further, the process of converting the image into gray scale may be executed at any point in the process in the face detection unit 3 or the feature point setting process.

  Next, the feature extraction unit 4 sets the positions of a plurality of feature points based on the positions of the detected facial organs (hereinafter referred to as “gaze points”: points indicating both eyes and mouth, for example). The feature extraction unit 4 sets feature points more densely as it is closer to the gazing point and coarser as it is farther from the gazing point.

  FIG. 2A is a diagram illustrating the face of the subject detected by the face detection unit 3. FIG. 2B is a diagram illustrating an example of a plurality of feature points set by the feature point setting process. In FIG. 2B, a black circle indicates a gazing point, and a hatched circle indicates a feature point set based on the gazing point. In the feature amount acquisition process described below, a gazing point may be handled as a feature point.

  Such feature point setting processing can be realized, for example, by applying Retina sampling described in the following paper.

F. SmeraldiandJ. Bigun, “Facial features detection by saccadic exploration of
the Gabordecomposition ”, International Conference on Image Processing, ICIP-98, Chicago, October 4-7, volume 3, pages 163-167, 1998.

<Feature acquisition processing>
In the feature amount acquisition process, the feature extraction unit 4 convolves a Gabor filter with each feature point set by the feature point setting process. That is, the feature extraction unit 4 performs Gabor Wavelets Transformation (GWT) for each feature point. FIG. 3 is an example (real part) of a Gabor filter used in the feature amount acquisition process. The feature extraction unit 4 obtains the periodicity and directionality of the light and dark features around the feature points as feature amounts by convolving a plurality of Gabor filters whose resolution and orientation are changed as shown in FIG.

Equation 1 is an expression representing a Gabor filter. In the use of the Gabor filter, by changing the values of k and θ in the equation, it is possible to acquire arbitrary periodicity and directionality as feature quantities from the density features.

  The feature extraction unit 4 passes the feature amount of each feature point obtained by the feature amount acquisition process to the attribute estimation unit 5. Note that the feature extraction unit 4 may be configured to perform processing on all faces that satisfy a predetermined condition among the faces detected by the face detection unit 3. The predetermined condition is, for example, a condition such as a face having a predetermined size or more, a face at a predetermined position (for example, an area at the center of the image), and a face having a predetermined direction (for example, facing the front).

[Attribute estimation part]
The attribute estimation unit 5 estimates the attribute information of the subject detected by the face detection unit 3 based on the feature amount of each feature point received from the feature extraction unit 4. The attribute information is information related to the person, and includes, for example, race, age, gender, and the like as items. The attribute estimation unit 5 estimates the attribute information of the subject by inputting the feature amount of each feature point to a pattern recognition discriminator for which learning processing has been completed in advance. The attribute estimation unit 5 performs processing using a support vector machine (SVM) as a pattern recognition classifier. Hereinafter, the support vector machine will be described by taking as an example the case of estimating the race among the attribute information.

The support vector machine is a pattern recognition method using a hyperplane passing through the middle of learning data belonging to two classes as an identification boundary surface. The discriminator in the support vector machine uses the discriminant function shown in Equation 2 to estimate which of the two classes the input data (here, feature quantities at all feature points) belongs.

In Equation 2, l indicates the number of learning data selected by the learning process, that is, the number of learning data used in the attribute estimation process by the attribute estimation unit 5. αi represents a Lagrange multiplier. xi and yi indicate learning data. yi has either “−1” or “1”, and indicates whether xi belongs to one of the two classes. b represents a bias term, that is, a parameter. These values are determined by the learning process, and the attribute estimation unit 5 stores the result of the learning process.

In Equation 2, K represents a kernel function. A nonlinear extension of the support vector machine has been proposed that uses a kernel function to map input data to higher dimensions in a non-linear manner, which makes it possible to construct classifiers that are more effective for real problems. . Typical examples of such a kernel function include a polynomial kernel (see Equation 3) and a Gaussian kernel (see Equation 4). In the attribute estimation unit 5, any kernel function may be applied.

  The support vector machine is a learning method that constitutes a discriminator for discriminating two classes. In order to discriminate (estimate) a plurality of (three or more) races, it is necessary to combine a plurality of support vector machines. The attribute estimation unit 5 implements multi-clustering using a support vector machine by applying a binary tree search. FIG. 4 is a diagram illustrating an example of a binary tree search applied to the race estimation process in the attribute estimation unit 5. Here, processing for estimating whether the race of the subject is a caucasian, a negroid, or a mongoloid will be described as an example. Other races may be further included as candidates according to the design.

  The attribute estimation unit 5 first estimates whether or not the subject person is a negroid using a support vector machine. When it is estimated that the subject person is not a negroid, the attribute estimation unit 5 estimates whether the subject person is a mongoloid (or a caucasian). The attribute estimation unit 5 similarly performs an estimation process for each item by performing a binary tree search for items of attribute information other than race, that is, age and sex. Then, the attribute estimation unit 5 passes the estimated attribute information to the attribute information determination unit 6. The attribute estimation unit 5 may be configured to perform processing on all faces that satisfy a predetermined condition among the faces detected by the face detection unit 3. Moreover, the attribute estimation part 5 performs estimation by performing the same process also about attribute information other than race, for example, sex and age.

[Attribute Information Determination Section]
When a plurality of faces are detected, attribute information is obtained for each face. When a plurality of attribute information is obtained in this way, the values of the items included in the respective attribute information may be different. In such a case, the value of each item of the attribute information output to the output control unit 7 needs to be determined as one. For this reason, the attribute information determination unit 6 determines which value to output for each item of the attribute information by performing the attribute information determination process. Hereinafter, a specific example of attribute information determination processing will be described.

First, the attribute information determination unit 6 generates a set for each face having each value for each item. Next, the attribute information determination unit 6 acquires the attention level for each set. The attention level is a value determined based on face information of faces included in each set. For example, the degree of attention is determined based on the number of faces included in each set, the face size, the face orientation, the face position, and the like. Then, the attribute information determination unit 6 determines the value in the set with the highest degree of attention as the value of the item. The attribute information determination unit 6 generates one attribute information by determining a value for each item, and passes the attribute information to the output control unit 7.

  FIG. 5 is a table showing an example of attribute information estimated by the attribute estimation unit 5. Hereinafter, as a specific example of the attention level, it is assumed that the higher the number of faces, the higher the attention level, and a specific example of attribute information determination processing will be described using the example shown in FIG. The attribute information determination unit 6 generates a set for each value of each item of race, age, and sex. The Caucasian set is (A), the Neggloid set is (B, E), and the Mongoloid set is (C, D, F). Similarly, the set of 10-20 is (E), the set of 20-30 is (A, B, C, D, F), the set of men is (A, F), the set of women is (B, C, D, E). Next, the attribute information determination unit 6 acquires the attention level of each set. In this case, since the attention degree of each set is obtained as the number of faces included in the set, (Caucasoid, Negroid, Mongoloid) = (1,2,3), (10-20, 20-30) = ( 1,5), (male, female) = (2,4). As a result, the attribute information determination unit 6 generates attribute information (Mongoloid, 20 to 30, female) and passes this attribute information to the output control unit 7. When only one face is detected, the attribute information determination unit 6 may pass the attribute information regarding the face to the output control unit 7 without executing the attribute information determination process.

(Output control unit)
The output control unit 7 controls the contents output to the output unit 8 based on the attribute information passed from the attribute information determination unit 6. The output control unit 7 controls the display and sound output from the output unit 8 according to the age included in the attribute information. For example, when the age is from 0 to 10, the displayed kanji may be controlled so that the furigana is added, may be controlled not to display using the kanji, You may comprise so that it may add and display. For example, when the age is 60 or more, the size of the displayed character may be controlled to be displayed larger than the normal time, or the volume of the output voice may be larger than the normal time. It may be controlled.

  Further, the output control unit 7 may be configured to control the color and design of displayed characters, the sex of output audio, and the like according to the gender included in the attribute information.

  Further, the output control unit 7 may select language candidates to be output to the output unit 8 according to the race included in the attribute information, for example. A language candidate is a candidate for selecting a language that the user desires to display or use. For example, language candidates corresponding to Mongoloid are Korean, Japanese, Thai, Chinese, Taiwanese, Mandarin, Arabic, Mongolian, Lao and the like. For example, language candidates corresponding to the Caucasian are English, French, German, Italian, Spanish, Dutch, Russian, Portuguese, Greek, and the like. For example, language candidates corresponding to negroid are Arabic, Portuguese, English, French, Swahili, and the like. In this case, the language to be used is selected from language candidates by the user via an input unit (not shown). Then, the output control unit 7 controls the output unit 8 to perform output using the selected language used. Further, the output control unit 7 may be configured to determine the language used regardless of the selection by the user, and to control the output unit 8 so as to perform output using the language used.

[Output section]
The output unit 8 is configured using a display device such as a CRT (Cathode Ray Tube) or a liquid crystal display, an audio output device such as a speaker, or the like. The output unit 8 may be configured using a touch panel or the like. The output unit 8 performs display of characters and images, audio output, and the like according to the contents of control by the output control unit 7.

[Operation example]
FIG. 6 is a flowchart illustrating an operation example of the terminal device 1. An operation example of the terminal device 1 will be described with reference to FIG. First, when the imaging unit 2 captures an image of the user (S01), the face detection unit 3 executes face detection processing on the captured image to detect the user's face (S02). Next, the feature extraction unit 4 and the attribute estimation unit 5 perform attribute estimation processing on the face detected by the face detection processing (S03).

  FIG. 7 is a flowchart illustrating an example of attribute estimation processing. An example of attribute estimation processing will be described with reference to FIG. When the attribute estimation process starts, the feature extraction unit 4 detects a gaze point in the face detected by the face detection unit 3, that is, a facial organ (S08). Next, the feature extraction unit 4 performs pre-processing based on the position of the point of sight on the detected face image (S09), and sets feature points (S10). And the feature extraction part 4 acquires a feature-value by convolving a Gabor filter with respect to each feature point (S11).

  When the feature extraction unit 4 acquires the feature amount, the attribute estimation unit 5 estimates the attribute (race, age, sex, etc.) of the subject by using the support vector machine with the acquired feature amount as an input (S12). . And the attribute estimation part 5 outputs the result of attribute estimation (S13).

  The processes after S03 will be described with reference to FIG. After the attribute estimation process (S03), the attribute information determination unit 6 determines the subsequent process according to the number of faces detected by the face detection unit 3. If the number of faces is 1 or less (S04: <= 1), the attribute information determination unit 6 notifies the output control unit 7 to that effect. Thereafter, the output control unit 7 performs output based on the attribute information of the face when one face is detected. For the case where no face is detected when the face is not detected. The output unit 8 is controlled so as to perform the output (for example, output performed at normal volume or normal character size, or output that displays all language candidates) (S06). . And the output part 8 performs an output according to the control content (S07).

  Returning to the description of the process in S04. When the number of faces is greater than 1 (S04:> 1), the attribute information determination unit 6 executes an attribute information determination process (S05). FIG. 8 is a flowchart illustrating an example of attribute information determination processing. A processing example of attribute information determination processing will be described with reference to FIG. When the attribute information determination process starts, the attribute information determination unit 6 acquires a set in each item (S14). Next, the attribute information determination unit 6 acquires the attention level in each set (S15). Then, the value of each item is determined according to the degree of attention in each set (S16). For example, the attribute information determination unit 6 may determine the value of the set having the highest degree of attention as the value of the item. And the attribute information determination part 6 produces | generates the attribute information which has the value determined about each item, and outputs the attribute information (S17).

  The processes after S05 will be described with reference to FIG. After the attribute information determination process (S05), the output control unit 7 controls the output unit 8 to perform output based on the attribute information output as a result of the attribute information determination process (S06). And the output part 8 performs an output according to the control content (S07).

[Action / Effect]
The terminal device 1 estimates user attribute information from the user's face image, and controls the output content based on the attribute information that is the estimation result. For example, the terminal device 1 may be configured to estimate a user's race as attribute information and present a user's language selection candidates (language candidates) estimated from the race. In this case, the user can select a language to be used from language candidates narrowed down to some extent by race estimation processing, output control processing, and the like, not from many language candidates. Therefore, it is possible to reduce the user's labor required for selecting the language to be used. For example, the terminal device 1 estimates the user's age as attribute information,
You may comprise so that the output according to the state of the user estimated from the age may be performed. Specifically, when the estimated age is elderly, the displayed characters may be controlled to be large, or may be controlled to increase the output volume. In addition, when the estimated age is young, it may be controlled not to use difficult kanji for the displayed characters, or may be controlled to shake the furigana. In this case, the user can receive output corresponding to different physical abilities and intellectual abilities according to the age, and the convenience of the user interface is improved. Further, for example, the terminal device 1 may be configured to estimate the user's gender as the attribute information and perform output according to the user's preference estimated from the gender. In this case, it is possible to receive an output that is more familiar to the user.

  Moreover, the terminal device 1 estimates a user's attribute from a user's face image. For this reason, the terminal device 1 can be installed in a noisy place and the like, and can be operated effectively in such a place.

  Further, in the terminal device 1, the user's attribute is estimated without using a recording medium in which attribute information or the like is registered. For this reason, when operating the terminal device 1, the user can be provided with a user interface corresponding to his / her attribute information without using a recording medium in which his / her attribute information is recorded.

[Modification]
The attribute information determination unit 6 may be configured to perform the determination in S04 using the number of faces subjected to the attribute estimation process instead of the number of faces detected by the face detection unit 3.

  Moreover, the output control part 7 may be comprised so that the contents, such as an output sentence, a word, and an image, may be controlled according to attribute information. For example, the output control unit 7 may perform control so as to output contents that are easy to understand or understand in accordance with the age of the user. For example, it is clear that there are differences in familiar names and understandable words depending on the age. Also, for example, in tourist information, the contents that the user would like to be guided by age and gender are different. Therefore, by performing such control, the user can obtain more effective information.

It is a figure which shows the functional block of a terminal device. It is a figure which shows the example of a setting of a feature point. It is a figure which shows the example of a Gabor filter. It is a figure which shows the example of a binary tree search. It is a table | surface which shows the example of the attribute information estimated by the attribute estimation part. It is a flowchart which shows the operation example of a terminal device. It is a flowchart which shows the operation example in an attribute estimation process. It is a flowchart which shows the operation example in an attribute information determination process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Terminal device 2 Imaging part 3 Face detection part 4 Feature extraction part 5 Attribute estimation part 6 Attribute information determination part 7 Output control part 8 Output part

Claims (7)

Imaging means for imaging a user's face;
Detecting means for detecting a user's face from an image captured by the imaging means;
Attribute estimation means for estimating user attributes from the face image detected by the detection means;
Output control means for controlling the content and mode of the output according to the estimation result by the attribute estimation means;
A terminal device comprising: output means for outputting according to the control content of the output control means. The attribute includes information about the race of the user,
The terminal device according to claim 1, wherein the output control unit controls the output unit to perform output in a language corresponding to the estimated race. The attribute includes information about the race of the user,
The output control means controls the output means to output a plurality of language candidates according to the estimated race,
The terminal device further includes a selection unit for the user to select a language used in the output unit from the plurality of language candidates.
The terminal device according to claim 1, wherein the output control unit controls the output unit to perform output in the selected language after the language is selected through the selection unit. The attribute includes information about the user's age,
When the estimated age can be determined to be old, the output control means displays the displayed characters larger than normal and / or outputs the output voice at a volume higher than normal. The terminal device according to claim 1, wherein the output unit is controlled to do so. When the estimation processing for a plurality of faces is performed by the attribute estimation unit, the output control unit further includes an estimation result selection unit that selects which of the plurality of types of estimation results should be used for processing. ,
The terminal device according to claim 1, wherein the output control unit performs processing using the estimation result selected by the estimation result selection unit. Imaging a user's face;
Detecting a user's face from the captured image;
Estimating user attributes from detected face images;
Controlling the content and mode of the output according to the estimation result;
A program for causing an information processing apparatus to execute the step of outputting according to the control content. An information processing device imaging a user's face;
An information processing device detecting a user's face from the captured image;
An information processing device estimating a user attribute from the detected face image;
An information processing device controlling the content and mode of output according to the estimation result; and
A method in which an information processing apparatus performs output in accordance with the control content.

Images