Apparatus for collecting preference information, and method for collecting preference information

Abstract

To more accurately grasp taste information for which many users prefer most.SOLUTION: A taste information collection device comprises: an image display unit on which images are displayed; an input unit which inputs a selection operation for a selected image selected by a user; a line-of-sight detection unit which detects a line of sight of the user when the user gazes at the image display unit; a gaze area determination unit which determines an area that user gases at for the longest time during a predetermined period of time, and outputs an image based on this area, as an image of interest; and a determination unit which determines whether the selected image is identical with the image of interest.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a preference information collection device and a preference information collection method.

Background Art

[0002] Conventionally, a list or icon related to content is displayed on a screen, and by detecting the selection status of a user with respect to an image including these lists or icons, characteristic behavior of the user is extracted to collect the user's preference information (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, with the conventional technology, although preference information selected based on the result of the user's thinking can be collected, preference information intuitively selected by the user cannot be obtained. For this reason, it has been difficult to more accurately grasp the preference information that many users feel is most preferable.

[0005] The present invention has been made in view of the above, and an object thereof is to more accurately grasp the preference information that many users feel is most preferable.

Means for Solving the Problems

[0006] In order to solve the above-described problems and achieve the object, a preference information collection device according to the present invention ​​​​​An image display unit that displays images, and an input unit that allows the user to input selection operations for the selected image. A power unit and a gaze detection unit that detects the user's gaze when the user is looking at the image display unit, The system determines the area where the user's gaze remained for the longest period of time and displays an image based on this area. A gaze region determination unit outputs an image as an eye image, and determines whether the selected image and the object of attention match. It comprises a determination unit and

[0007] Furthermore, the method for collecting preference information according to the present invention includes the step of displaying an image, The steps involve the user inputting a selection operation for the selected image, and the user displaying the image. The steps include detecting the user's gaze when they are looking at a particular part, and determining when the user's gaze is at its most important point within a predetermined time. This step involves identifying regions that have existed for a long time and outputting an image based on these regions as the image of interest. The system includes the step of determining whether the selected image and the image of interest match. [Effects of the Invention]

[0008] According to the present invention, it is possible to more accurately grasp preference information that many users find most desirable. It is possible. [Brief explanation of the drawing]

[0009] [Figure 1] Figure 1 is a block diagram showing an example configuration of a preference information collection device according to this embodiment. [Figure 2] Figure 2 is a flowchart showing the processing flow of user preference surveys in the preference information collection device according to this embodiment. [Figure 3] Figure 3 is a diagram illustrating the common feature points between the selected image and the image of interest. [Figure 4] Figure 4 is a flowchart showing the processing flow for identifying the image that many users find most desirable in the preference information collection device according to this embodiment.

Best Mode for Carrying Out the Invention

[0010] Hereinafter, with reference to the accompanying drawings, embodiments of a preference information collection device and a preference information collection method according to the present invention will be described in detail. Note that the present invention is not limited by the following embodiments. No.

[0011] FIG. 1 is a block diagram showing a configuration example of a preference information collection device according to the present embodiment. The preference information collection device 10 investigates the user's preference images (preferences) from the images (selected images) actually selected by the user from a plurality of comparison images displayed on the screen and the images (gazed images) on which the user has gazed due to the movement of the user's line of sight, and more accurately grasps the preference information that many users feel is the most preferable. The preference information collection device 10 is an information processing device such as a personal computer, and includes an arithmetic processing device (control device) configured by a CPU (Central Processing Unit) or the like. The preference information collection device 10 loads the stored program into the memory and executes the instructions included in the program. The preference information collection device 10 includes an internal memory not shown in the figure, and the internal memory is used for temporarily storing data such as programs in the preference information collection device 10. As shown in FIG. 1, the preference information collection device 10 includes an image display unit 11, a line-of-sight detection unit 12, an input unit 13, a gaze area determination unit 14, a user information unit 15, a determination unit 16, a tabulation unit 17, an analysis unit 18, an image control unit 19, and an image storage unit 20.

[0012] The preference information collection device 10 is an information processing device such as a personal computer, and includes an arithmetic processing device (control device) configured by a CPU (Central Processing Unit) or the like. The preference information collection device 10 loads the stored program into the memory and executes the instructions included in the program. The preference information collection device 10 includes an internal memory not shown in the figure, and the internal memory is used for temporarily storing data such as programs in the preference information collection device 10. As shown in FIG. 1, the preference information collection device 10 includes an image display unit 11, a line-of-sight detection unit 12, an input unit 13, a gaze area determination unit 14, a user information unit 15, a determination unit 16, a tabulation unit 17, an analysis unit 18, an image control unit 19, and an image storage unit 20.

[0013] ​​​​​​​​​The image display unit 11 is a display device provided in the preference information collection device 10. 1 is, for example, a liquid crystal display (LCD: Liquid Crystal Display). ay) or OLED (Electro-Luminescence) displays, etc. The display includes a multi-image display unit 11 which receives multiple images via the image control unit 19. The numerical comparison images are displayed in predetermined areas. In the example in Figure 1, the image display unit 1 1 is divided into four regions 30A to 30D, and each region 30A to 30D is Comparison images (images) 31A to 31D of a given product (e.g., passenger car) are displayed. Images 31A-31D are still images, but they may also be videos. Comparison images 31A-31D are: This represents the image of the product, showing the characteristic features of the product as a whole or its parts, such as color and shape. It includes this feature. This feature is that when the user selects comparison images 31A to 31D, the user This is the part that will pique their interest and attention. This characteristic will be discussed later. Figure 1 shows the product While I used a passenger car as an example, it is not limited to this as long as it expresses the distinctive features of shape and color. No.

[0014] The gaze detection unit 12 is positioned adjacent to the image display unit 11, and when the user gazes at the image display unit 11... When viewed, it detects the user's gaze. The gaze detection unit 12 uses multiple (for example, two) cameras 1 It has 2A, and by capturing the user's eyes with this camera 12A, the direction of the user's gaze can be determined. (Motion) is detected. Information regarding the detected gaze direction is output to the gaze area determination unit 14. ru.

[0015] The input unit 13 is an input device that inputs various types of information to the preference information collection device 10. 13 can use, for example, a keyboard or mouse. In this embodiment, an image Button switches (not shown) are superimposed on each of the areas 30A to 30D provided on the display unit 11. It has been done. Then, the user uses the mouse to compare images 31A to 31D. The image that is considered most preferable can be selected by performing an input operation on that image (for example, image 3 Image 1B) is selected as the selected image. Information regarding the selected image is output to the determination unit 16. To be empowered.

[0016] The gaze area determination unit 14 determines the gaze area based on the user's gaze detected by the gaze detection unit 12. The time during which the line of sight was present in each region 30A to 30D of the image display unit 11 is measured. The gaze area determination unit 14 determines which of the areas 30A to 30D of the image display unit 11 is used for a predetermined time. The gaze region determination unit 14 determines the region where Za's gaze was directed for the longest period (e.g., region 30C). The image 31C displayed in the determined region 30C is acquired as the image of interest, and this image of interest Image information is output to the determination unit 16. In addition, the gaze area determination unit 14 determines each area 30A Information regarding the time when each line of sight existed at ~30D is output to the analysis unit 18. Here, For a predetermined time, for example, when the comparison images 31A to 31D are displayed in the areas 30A to 30D of the image display unit 11 This refers to the time from when the image is displayed until the user makes a selection or inputs an image. The area where the user's gaze remained for the longest period of time is the area that the user was fixated on within the specified time period mentioned above. This refers to the region where the total time spent is the longest, for example, when comparing images from two regions alternately. It may be a discontinuous period of time. Also, the presence of the user's gaze in that area means that the user This refers to the observation of the subject's gaze continuously for a certain period of time (e.g., 0.1 seconds) or longer within that area. This eliminates situations where the area is accidentally seen by the user, regardless of their intention. It can accurately determine the area (image) that the user is focusing on.

[0017] The user information unit 15 collects user information such as gender, age, preferences (areas of interest), etc. Obtain information about gender (attribute information). Information about the user's attributes is obtained from the multiple methods described above. When performing a series of operations to select an image from a comparison image, input is used, for example, via the input unit 13. The acquired user attribute information is output to the analysis unit 18.

[0018] The determination unit 16 uses the selected image selected via the input unit 13 and the gaze area determination unit 14 to determine the selected image. The determination unit 16 determines whether the outputted image of interest matches or not. The report is output to the aggregation unit 17.

[0019] The aggregation unit 17 counts (aggregates) the images based on the input judgment result. If it is determined that the selected image and the image of interest match, the aggregation unit 17 will use this selected image. This will be counted as a representative image of interest to the user. Also, if the selected image and the featured image do not match... If it is determined that the image does not match the selected image, the aggregation unit 17 will determine the image of interest that does not match the selected image. Count as an image, and count as a logical image an image among the selected images that does not match the image of interest. The aggregation unit 17 also aggregates the number of representative images, sensory images, and logical images for each user. The aggregated results are then output to the analysis unit 18. Here, the representative image is selected from among multiple comparison images. These are images selected by users as being of interest, and are the most frequently chosen by multiple users. The representative image selected will be treated as the recommended image, as described later.

[0020] The analysis unit 18 includes representative images, sensory images, and logical images counted by the aggregation unit 17. Information regarding the total number of each image, and information regarding the user's attributes obtained from the user information unit 15. Based on the report, statistical information tailored to user attributes is generated. This statistical information includes the user's attributes. Representative images in preference surveys, management numbers for these representative images, number of representative images, logical image This includes the number of times and the time that a line of sight existed in each region 30A to 30D. The generated statistical information is output to the image control unit 19. The analysis unit 18 also analyzes multiple users Based on the statistical information, the representative image with the highest count will be selected as the recommended image. This represents the images (preference information) that many users find most appealing. In this scenario, the image selected by the user as being of interest from among multiple comparison images is designated as the representative image. Because it is determined to be an image, within the range categorized by the specified user attributes, multiple The representative image that was most frequently selected by a number of users is the one that these users found most appealing. It can be recognized as a recommended image.

[0021] The image control unit 19 displays multiple comparison images of the target product in a predetermined area 3 of the image display unit 11. Control is performed to display from 0A to 30D. These comparison images are pre-set for each product. It is configured. Furthermore, depending on the user's image selection, a new set of comparison images is created. This may also be the case. Specifically, the image control unit 19 determines that the selected image and the image of interest do not match. If this occurs, pre-defined feature points are extracted from the selected image and the image of interest, and matching feature points are selected. Select new comparison images and display these newly selected images on the image display unit 11. The image that was previously displayed is replaced and shown on the image display unit 11.

[0022] The image storage unit 20 stores multiple comparison images of the target product, and image control The image selected by the control of unit 19 is output. The image storage unit 20 is controlled by the image control unit 1 Control 9 stores statistical information tailored to user attributes.

[0023] Next, the processing flow of the preference information collection device 10 will be explained. Figure 2 shows the flow of this embodiment. A flowchart showing the processing flow of user preference surveys in the preference information collection device 10. Figure 3 is a diagram illustrating the common feature points between the selected image and the image of interest.

[0024] The preference information collection device 10 displays the comparison images that are the subject of the preference survey in each area of ​​the image display unit 11. Display (Step S1). Specifically, the preference information collection device 10 is controlled by the image control unit 19. Then, multiple comparison images related to the target product are read from the image storage unit 20, and these comparison images The image is displayed in each area of ​​the image display unit 11.

[0025] Next, the preference information collection device 10 inputs the selected image chosen by the user (step S2 Specifically, the preference information collection device 10, based on the user's operation via the input unit 13, From among multiple comparison images, the user selects the image they find most preferable as the selected image. The selected image is input. Information regarding this selected image is output to the determination unit 16.

[0026] Next, the preference information collection device 10 acquires user attribute information (step S3). The user information unit 15 of the information gathering device 10, based on the user's operation via the input unit 13, We obtain user attribute information such as gender, age, and preferences (hobbies or areas of interest). In the example in Figure 2, the configuration is such that user attribute information is obtained after the selected image is entered. However, the timing for obtaining user attribute information is before the selected image is entered. good.

[0027] Next, the preference information collection device 10 acquires the user's attention image based on gaze detection ( Step S4). Specifically, the preference information collection device 10 uses the gaze area determination unit 14 to determine the user Based on the user's gaze, within the regions 30A to 30D of the image display unit 11, the user's gaze at a predetermined time Determine the region where the line is longest (for example, region 30C). Then, determine the determined region 3 The image 31C displayed at 0C is acquired as the image of interest. In this case, the predetermined time is used for the image display. From the moment the comparison images 31A to 31D are displayed in the regions 30A to 30D of the display unit 11, the user can access the information. This is the time until the input operation for the selected image is performed. Also, the acquired focus image is... The relevant information is output to the determination unit 16.

[0028] Next, the preference information collection device 10 determines whether the selected image and the image of interest match. Step S5). Specifically, the preference information collection device 10 uses the information input to the determination unit 16 to determine Based on this, the selected image chosen by the user and the image that the user focused on for the longest time are... Determine whether they match or not. In this determination, if the selected image and the image of interest match ( In step S5 (Yes), the preference information collection device 10 proceeds to step S6.

[0029] In contrast, if the selected image and the image of interest do not match (Step S5; No), The preference information collection device 10 aggregates the unmatched selected images as logical images (step S9). The preference information collection device 10 counts the number of logical images using the aggregation unit 17. A logical image is an image among the selected images that does not match the image of interest. Also, among the images of interest... Images that do not match the selected image may be counted as sensory images.

[0030] Next, the preference information collection device 10 extracts matching feature points between the selected image and the image of interest (S Step S10). Specifically, in Figure 3, select the image 31B displayed in region 30B. In this embodiment, the image 31C displayed in region 30C will be described as the image of interest. These images 31B and 31C have pre-defined characteristic points of the product (passenger car). For passenger cars, this would include the overall shape of the body, the color of the body, the shape of the headlights, and the shape of the taillights. Features include the shape of the doors, the shape of the windows (side windows), and the shape of the tire wheels. These are the settings. These feature points are just examples, and even if you set feature points in other locations... Yes, the number of feature points may be increased or decreased. The preference information collection device 10 is controlled by the image control unit 19. The image control unit 19 extracts matching feature points from the selected image and the focus image. By performing image analysis with the image, the shapes of the windows (side windows) 32B and 32C can be determined. These are extracted as key feature points.

[0031] Next, the preference information collection device 10 selects an image having feature points that match the extracted feature points. Select (step S11). That is, the preference information collection device 10 is controlled by the image control unit 19. From among the images of the product (passenger car) stored in the image storage unit 20, a selected image and an image of interest are selected. The image has windows with the same shape as the feature points 32B and 32C (matching feature points). Select two images. Then, the preference information collection device 10 collects the selected images and the attention image and the selected Adding the two images created, a new comparison image is created, and the image displayed on the image display unit 11 is placed next to it. The display is changed, and the process returns to step S1 to execute the repeating process. The preference information collection device 10 selects four images that have feature points that match the extracted feature points. You may select four images and use them as new comparison images.

[0032] In the determination in step S5 described above, if the selected image and the image of interest match (step S5; Yes), the preference information collection device 10 selects matching images that are of interest to the user. The images are tallied as representative images (Step S6). The representative images are multiple (four) comparison images. This is the image that the user found most interesting among the images. In this embodiment, the image actually selected by the user If the selected image matches the image the user has looked at the longest, the selected image will be used as the representative image. To achieve this, it is possible to eliminate cases where the selected image was accidentally selected, and the user It is possible to accurately compile representative images of interest.

[0033] Next, the preference information collection device 10 determines whether or not there is an additional investigation (step S7). Here, the additional investigation involves another comparison within the same product (passenger car) in the preference survey mentioned above. This is done using images and comparison images of other products (such as motorcycles). If further investigation is needed (Step S7; Yes), return the process to Step S1. Execute the process repeatedly.

[0034] Furthermore, if no further investigation is needed (Step S7; No), the preference information collection device 10 will... - Output statistical information tailored to the user's attributes (Step S8) and terminate. Preference Information Collection Device 1 0 is determined by the analysis unit 18, which associates the user's attribute information with the user's survey results, The system generates statistical information tailored to the attributes of the image and outputs this statistical information to the image control unit 19. The control unit 19 stores this statistical information in the image storage unit 20 and terminates processing. This includes various representative images used in user preference surveys, management numbers for representative images, and the number of representative images. The number of logical images, and the time that a line of sight existed in each region 30A to 30D, at least include.

[0035] Next, based on statistical information from multiple users, we will identify the image that many users find most appealing (preference information). The following describes the process for obtaining the information. Figure 4 shows the preference information collection according to this embodiment. This flowchart illustrates the process of identifying the image that most users of the device find most desirable. —This is a chart.

[0036] First, the preference information collection device 10 inputs statistical information for each user (step S21). Specifically, the image control unit 19 reads the statistical information stored in the image storage unit 20 for each user. Then, statistical information for each user is input into the analysis unit 18.

[0037] Next, the preference information collection device 10, based on the statistical information of each user that has been input, generates a logical image Determine whether or not there were users whose number exceeded a predetermined threshold (step S22). This determines whether the number of logical images counted for the same user exceeds a predetermined threshold. Determine. This predetermined threshold is set as appropriate, for example, a series of adjustments to the user. It is preferable to set the quantity to 1 / 3 to 1 / 2 of the quantity being examined. In this determination, preference information is collected. The collection device 10, through its analysis unit 18, determines if there are no users whose number of logical images exceeds a predetermined threshold. If it is determined that (Step S22; No), the process proceeds to Step S25.

[0038] On the other hand, the preference information collection device 10 determines that there was a user whose number of logical images exceeded a predetermined threshold. If determined (Step S22; Yes), the analysis unit 18 will analyze this same user If, during the time it takes to generate the statistical information, there are areas where the user's gaze has not been detected, Determine whether or not (step S23). Generation of statistical information for the same user is completed. The period until that point refers to all of the featured images output for the same user. This refers to the period until a predetermined time has elapsed. In this case, the analysis unit 18 analyzes each region 30A to 30D Based on the time that each line of sight was present, we determine whether or not there were any areas where no line of sight was detected. The device makes a determination. In this determination, the preference information collection device 10 collects statistical information about the same user. It was determined that there were no areas where the user's gaze was not detected during the time the generation process was completed. If this occurs (Step S23; No), the process proceeds to Step S25.

[0039] Meanwhile, the preference information collection device 10 continues until the generation of statistical information about the same user is completed. If it is determined that there was an area in which the user's gaze was not detected during that time (Step S2 3; Yes) To do so, discontinue the use of the statistics for the user in question (step S24). That is, if the number of logical images counted for the same user exceeds a predetermined threshold, and the same All predetermined time periods corresponding to all featured images output for a single user have elapsed. If there is an area in between where the user's gaze has not been detected, then the user's preferences It can be determined that the investigation is being conducted haphazardly (fraudulently). Since the statistics do not accurately reflect user preferences, this user statistical information is... Using it to identify the most preferred image among multiple users can actually lead to noise and... There is a risk that this will happen. For this reason, in this embodiment, the preference survey is not conducted correctly. By discontinuing the use of user statistics for users who appear to be such, we can more accurately understand the preferences of multiple users. It can be understood.

[0040] Next, the preference information collection device 10 aggregates representative images for each user (step S25), Based on user attributes, the most frequently selected representative image is chosen as the recommended image, and this recommended image information is output. (Step S26). As described above, the representative image is the one that users find most interesting among the comparison images. The image has a distinctive character. The preference information collection device 10, through its analysis unit 18, analyzes the characteristics of numerous users. By aggregating the table images by user attributes (gender and age), the users can be categorized. In terms of attributes, it accurately analyzes and identifies recommended images, which are the images that the majority of users found most interesting. The analysis unit 18 of the preference information collection device 10 can make recommendations corresponding to user attributes. By outputting image information to, for example, an external management device (not shown), a large number of users can... It can also be proposed as a design image for the most suitable product that the user is interested in.

[0041] As described above, the preference information collection device 10 according to this embodiment has a plurality of images 31A to 31D that it has pre-processed Image display unit 11 to display in each of the designated areas 30A to 30D, and image display unit 11 Select the image selected by the user from among the multiple images 31A to 31D displayed. An input unit 13 for inputting data, and when the user gazes at the image display unit 11, the unit detects the user's line of sight. The gaze detection unit 12 emits an image, and within the region 30A to 30D of the image display unit 11, the user selects the desired direction for a predetermined time. The system determines the region where Za's gaze was directed for the longest period and outputs the image of this region as the image of interest. A visual area determination unit 14 and a determination unit 16 that determines whether the selected image and the image of interest match, If it is determined that the selected image and the featured image match, the selected image will be displayed to the user as an image of interest. It includes an aggregation unit 17 that counts as a table image. If the selected image matches the image the user has looked at the longest, the selected image will be used as the representative image. By doing so, it is possible to eliminate cases where the selected image was selected by mistake, It is possible to accurately collect representative images that users are interested in. Therefore, many users will find the most interesting. It becomes possible to more accurately understand images that people find appealing (preference information).

[0042] Furthermore, images 31A to 31D each have multiple feature points that are pre-defined as having a distinctive shape. If it is determined that the selected image and the featured image do not match, the feature points of the selected image and the featured image will be used. The feature points of the eye image are compared with the feature points of the eye image, and matching feature points are extracted. A new image having the specified characteristics is selected and displayed, replacing the image previously displayed on the image display unit 11. Because it includes an image control unit 19, it provides the user with an image containing feature points that the user has focused on. This allows for accurate aggregation of representative images that users are interested in.

[0043] Furthermore, the aggregation unit 17 counts representative images for each user. The most frequently viewed representative image in the table will be designated as the recommended image that multiple users find most preferable. Because it is equipped with an analysis unit 18, in the categorized user attributes, a large number of users are the most It can accurately identify recommended images that users are interested in, and these recommended images can be shared by many users. It can be proposed as the most interesting and optimal product design image.

[0044] Furthermore, the aggregation unit 17 determined that, for each user, the selected images did not match the image of interest. The images are counted as logical images, and the analysis unit 18 counts for the same user. If the number of logical images exceeds the threshold, and all featured images output for the same user are... Within the time elapsed for all corresponding predetermined periods, the user's gaze is detected in multiple areas. If there are areas where the recommendation is not made, the user's statistical information (representative image) will be used to determine the recommended image. Because it does not use, it is possible to eliminate information that does not accurately reflect the user's preferences, and many It allows for accurate understanding of user preferences.

[0045] The preference information collection device 10 according to this embodiment has been described above, but the above embodiment It may also be implemented in various other forms. Each component of the illustrated preference information collection device 10 The elements are functional concepts and do not necessarily have to be physically constructed as shown in the diagram. i. In other words, the specific form of each device is not limited to those shown in the illustration, and the processing load and usage of each device may vary. Depending on the circumstances, all or part of it may be functionally or physically dispersed in any unit. It's okay to merge them.

[0046] The configuration of the preference information collection device 10 is, for example, as software, loaded into memory. This is implemented by a program or the like. In the above embodiment, these hardware or software It was explained as a functional block realized through the cooperation of software. Functional blocks can be hardware only, software only, or a combination of both. It can be realized in various forms depending on the combination.

[0047] The above-mentioned components include those that can be easily conceived by a person skilled in the art, and those that are substantially the same. Furthermore, the above-described configurations can be combined as appropriate. Within the scope, various omissions, substitutions, or modifications of the configuration are possible. [Explanation of symbols]

[0048] 10. Preference Information Collection Device 11 Image display section 12 Eye-line detection unit 12A Camera 13 Input section 14 Gaze area determination unit 15 User Information Section 16 Judgment section 17. Aggregation Department 18 Analysis Department 19 Image Control Unit 20 Image storage unit 30A, 30B, 30C, 30D area Images 31A, 31B, 31C, 31D Window shapes (characteristic features) of 32B and 32C

Claims

1. An image display unit that displays images, An input section for inputting selection operations for the selected image chosen by the user, When the user gazes at the image display unit, the gaze detection unit detects the user's gaze. 、 The system determines the region where the user's gaze was directed for the longest period of time, and generates an image based on this region. A gaze area determination unit that outputs the image of interest, A determination unit that determines whether the selected image and the image of interest match, A preference information collection device equipped with the following features.

2. Each of the aforementioned images has multiple feature points, which are pre-defined as having a distinctive shape. If it is determined that the selected image and the image of interest do not match, The feature points of the selected image and the feature points of the image of interest are compared and matching feature points are extracted. A new image having feature points similar to those extracted above is selected and displayed on the image display unit. The preference information according to claim 1, comprising an image control unit that replaces and displays the previously shown image. Collection device.

3. A method for collecting preference information performed by a preference information collection device, The steps include displaying an image on the image display unit of the preference information collection device, The steps include: inputting a selection operation for the selected image chosen by the user, The steps include detecting the user's gaze when the user is looking at the image display unit, The system determines the region where the user's gaze was directed for the longest period of time, and generates an image based on this region. The steps include outputting the image as the image of interest, The steps include determining whether the selected image and the image of interest match, A method for collecting preference information that includes the following features.

Images