Makeup proposal method, makeup evaluation method, makeup support method, makeup proposal device, and makeup evaluation device

Abstract

To support makeup in consideration of a situation where a face applied with makeup is seen.SOLUTION: Provided is a makeup proposal method including acquiring information related to an observed situation of a subject face to be applied with makeup, and proposing makeup suitable for the observed situation on the basis of the information. The proposed makeup includes makeup providing a predetermined space frequency characteristic.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a makeup suggestion method, a makeup evaluation method, a makeup support method, a makeup suggestion device, and a makeup evaluation device. [Background technology]

[0002] In the field of beauty, various methods and devices for supporting users in applying makeup are known. For example, cited reference 1 discloses a technology that shows a user recommended makeup candidates and product information that matches the makeup candidates based on the user's interest level in facial features of a simulation image of a face with makeup applied. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6748981 Summary of the Invention [Problem to be solved by the invention]

[0004] There are various situations in which a user's made-up face can be seen. For example, there are situations in which the user meets someone in person and has a daily conversation, in which the user is seen by an observer at a close distance, and there are also situations in which the user gives a lecture in a large venue, in which the user is seen by an observer at a distance. There are also situations in which the user's face can be seen through a screen. Here, it is empirically understood that the way makeup is seen changes depending on the situation, and some users use different makeup depending on the situation in which the user's face will be seen. However, even if the user intends to apply makeup according to the situation, the effect of the makeup may not be very good. For example, even if the user intends to apply makeup to emphasize the eyes in order to make the eyes look larger even in a situation in which the user is seen from a distance, the effect of making the eyes larger may not be very good in reality. However, there has been no technology to obtain the user's desired makeup according to the situation in which the user's face will be seen.

[0005] In view of the above, an object of one embodiment of the present invention is to provide makeup application support that takes into consideration a situation in which a face with makeup applied is visible. [Means for solving the problem]

[0006] One aspect of the present invention may be a makeup suggestion method that includes acquiring information regarding the observation situation of the face of a subject to whom makeup is to be applied, and suggesting makeup suitable for the observation situation based on the information, wherein the suggested makeup includes makeup that results in a predetermined spatial frequency characteristic. Effect of the Invention

[0007] According to one aspect of the present invention, it is possible to provide makeup application assistance that takes into consideration the situation in which the face on which makeup is applied will be visible. [Brief description of the drawings]

[0008] [Figure 1] FIG. 2 is a flow diagram of a method for constructing a database used in an embodiment of the present invention. [Diagram 2] FIG. 1 is a flow diagram of a makeup suggestion method. [Diagram 3] FIG. 11 is a flow diagram of a specific example of a makeup suggestion method. [Figure 4] 4 is a diagram showing an example of a makeup pattern proposed by the makeup suggestion method of FIG. 3. [Diagram 5] FIG. 2 is a functional block diagram of the makeup suggestion device. [Figure 6] FIG. 1 is a flow diagram of a makeup evaluation method. [Figure 7] FIG. 1 is a flow diagram of a specific example of a makeup evaluation method. [Figure 8] FIG. 8 is a diagram showing an example of an evaluation result in the makeup evaluation method of FIG. 7. [Figure 9] FIG. 2 is a functional block diagram of the makeup evaluation device. [Figure 10] FIG. 2 is a diagram illustrating an example of a hardware configuration of a makeup suggestion device and a makeup evaluation device. [Figure 11]11A to 11C are diagrams illustrating the formation of partial images in an embodiment. [Figure 12] FIG. 11 is a power spectrum diagram of an image in an experimental example. [Figure 13] FIG. 13 is a diagram showing the relationship between the viewing distance and the amount of optical illusion for images in an experimental example. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] One embodiment of the present invention is a method for assisting with makeup. One embodiment is a method for suggesting makeup, comprising acquiring information about an observation state of a face of a person to whom makeup is to be applied, and proposing makeup suitable for the observation state based on the information, the suggested makeup including makeup that provides a predetermined spatial frequency characteristic. Another embodiment is a method for acquiring an image of a face of a person to whom makeup is applied, and evaluating the makeup for suitability with the observation state of the face of the person based on the spatial frequency characteristics of the made-up portion of the image.

[0010] In this specification, "makeup (also called makeup or makeup)" refers to a treatment or technique for changing the impression of the face by applying a specific material to the face. Also, "makeup product (also called makeup material or cosmetic material)" refers to a material used for the above-mentioned makeup, that is, a material for changing the impression of the face by applying it to the face (in many cases, by painting it on the face).

[0011] The makeup in this specification may include one or more of eye makeup and lip makeup. Eye makeup (also called eyebrow area makeup) is makeup applied to an area including the eyes and eyebrows, and is makeup that changes the impression of the eyes and / or eyebrows. For example, the upper and lower eyelids, eye holes, and areas called tear bags are also included in the area to which eye makeup is applied. Examples of makeup products used for eye makeup include eye shadow, eyeliner, mascara, false eyelashes, and eyebrows. Lip makeup (also called lip area makeup) is makeup applied to an area including the lips, and is makeup that changes the impression of the lips, mouth, or mouth area. Examples of makeup products used for lip makeup include lipstick and lip gloss. In either case, a makeup product for suppressing color such as a concealer can also be used. Among the above makeups, eye makeup is particularly one that greatly changes the impression of the face, so it is preferable that the makeup in this embodiment is eye makeup.

[0012] In this specification, the "face observation situation" refers to a situation in which the face of a subject with makeup applied is seen by an observer. For example, in a situation in which the subject is directly (realistically or actually) seen by an observer, it may be the distance between the subject and the observer (observation distance). Furthermore, the face observation situation may include factors that may affect the impression that the observer receives when seeing the subject's face, such as indoor or outdoor type, lighting, weather, the presence or absence of a shield between the subject and the observer, and the type of shield. Furthermore, in a case in which the subject is observed or seen by the observer through a display, that is, the subject's face is once photographed and the data is imported into the device, and an image generated from the data is displayed on the display, and the subject's face displayed on the display may be seen by the observer. Furthermore, in a situation in which the face is observed through a display, the face observation situation may also include the type of hardware used to capture and display the face, such as the type and performance of the camera, the performance of the display (including the resolution of the display and the magnification and reduction ratio of the face on the display), and the type of software used to capture and display the face.

[0013] Commonly assumed face observation situations include a situation in which the face is observed directly at a close distance (called "real close distance"), a situation in which the face is observed directly at a long distance (called "real long distance"), and a situation in which the face is observed through a display (called "display display"). The real close distance is a situation in which a person wearing makeup and an observer directly talk in normal voices, and the distance between the person wearing makeup and the observer can be about 0.4 m to 2 m. The real long distance is a situation in which a person wearing makeup is observed by an observer on stage, and the distance between the person wearing makeup and the observer can be more than 2 m. Note that the way in which the face is perceived, particularly the way in which the size of the eyes is perceived, which can be measured as the amount of optical illusion (described later), has the same tendency in both real long distance and display display, because in display display, the face is displayed smaller than the actual face and high spatial frequency components are lost, making it difficult to distinguish details. Therefore, the commonly assumed face observation situations described above may be classified into two: real close distance (observation situation I) and real long distance and display display (observation situation II).

[0014] Many users have traditionally changed their makeup depending on the situation in which their face is observed. For example, many users apply makeup that emphasizes their eyes, assuming a situation in which they will be seen by observers at a distance, such as when performing on stage or giving a lecture. However, makeup that the user believes to have a large effect of emphasizing the eyes may not have a significant effect of making the eyes larger. On the other hand, when attempting to apply makeup that emphasizes the eyes assuming a situation in which the user will be seen at close range (i.e., a normal observation situation in daily life), the desired effect of making the eyes look larger may be achieved, but the application of an excessively large amount of makeup product may result in an unnatural appearance.

[0015] Therefore, the inventors of the present invention have conducted extensive research into how makeup-induced illusionary effects, i.e., effects of changing the impression of facial features, such as making the eyes look larger, making the lips look larger, plumper, or thicker, differ depending on makeup patterns and observation conditions. As a result, the inventors have found that the spatial frequency characteristics of makeup patterns affect the illusionary effects in a given observation condition. For example, the inventors have found that makeup with a larger power spectrum value (intensity) in the high spatial frequency region is more likely to cause the illusionary effect to be perceived at close ranges, and makeup with a larger power spectrum value in the low spatial frequency region is more likely to cause the illusionary effect to be perceived at long ranges. In other words, makeup that has a large effect of increasing the spectral intensity (spectral signal intensity) in the low spatial frequency region in the spatial frequency power spectrum tends to have a large illusionary effect at long ranges, and makeup that has a large effect of increasing the spectral intensity (spectral signal intensity) in the high spatial frequency region in the spatial frequency power spectrum tends to have a large illusionary effect at close ranges. Also, for example, makeup with a relatively high proportion of high spatial frequency components in the spatial frequency power spectrum tends to have a greater illusionary effect at close range than makeup with a relatively low proportion of high spatial frequency components, and makeup with a relatively high proportion of low spatial frequency components in the spatial frequency power spectrum tends to have a greater illusionary effect at long range than makeup with a relatively low proportion of low spatial frequency components. This embodiment has been made based on such knowledge, and can propose makeup suitable for the observed situation according to the user's desire. Also, it can evaluate what kind of observed situation the makeup already applied by the user is suitable for, or whether the desired illusionary effect can be obtained in the observed situation assumed by the user.

[0016] <Database construction> In both the makeup suggestion method and the makeup evaluation method described above, it is preferable to build a database by acquiring the relationship between the spatial frequency characteristics of makeup and the amount of optical illusion for each observed situation. Alternatively, when the observed situation is an observed distance, a quantitative relationship may be obtained between the spatial frequency characteristics of makeup, the amount of optical illusion, and the observed distance. A method for building such a database will be described below, but the description is mainly based on an example of eye makeup.

[0017] A method for constructing such a database may include preparing a plurality of different makeup patterns, obtaining a plurality of face images with the different makeup patterns (S01), obtaining spatial frequency characteristics for each of the plurality of face images (S02), measuring the amount of optical illusion for each of the plurality of face images for different observation situations (S03), and determining the relationship between the spatial frequency characteristics of the face image and the amount of optical illusion for each observation situation (S04) (FIG. 1).

[0018] In this specification, makeup includes the type of makeup product used, as well as the makeup technique, such as the area to which makeup is applied, the degree to which makeup is applied (i.e., the thickness of the makeup), and the method of applying gradation of the makeup product (gradient, etc.). Thus, a makeup pattern may be formed by one type of makeup product, or may be formed by a combination of multiple types of makeup products.

[0019] In the face image acquisition (S01), a predetermined pattern of makeup may be actually applied to a target model, an image of the face may be photographed using an imaging device such as a camera, and then another pattern of makeup may be applied to the same target model and photographed repeatedly to acquire face images with a plurality of different makeup patterns. Alternatively, an image of the natural face of the target model (image without makeup) may be acquired, and various makeup patterns may be created for that image by image processing to acquire images with a plurality of different makeup patterns. Furthermore, a group of such face images with a plurality of different makeup patterns may be acquired for each of a plurality of target models.

[0020] Next, the spatial frequency characteristic is acquired (S02) for each of the multiple face images with different makeup patterns obtained in the step (S01) of acquiring face images. The spatial frequency characteristic can be acquired, for example, by performing a spectral analysis on the spatial frequency of each image. The result of the spectral analysis may be output as a power spectrum diagram. Then, based on the obtained power spectrum diagram in the form of a graph or data, a feature amount in a predetermined spatial frequency region of each image may be extracted. For example, one or more feature amounts of a representative value such as an average value or a median value of the power spectrum value (intensity) in a relatively high spatial frequency region (high spatial frequency region) and an integral value may be acquired. Alternatively or additionally, a feature amount similar to the above may be acquired in a relatively low spatial frequency region (high spatial frequency region). Alternatively, a ratio of the spatial frequency component in a predetermined region to the entire spatial frequency power spectrum, for example, a ratio of the integral value of the power spectrum value in a predetermined spatial frequency region to the total integral value of the power spectrum value, may be acquired as a feature amount.

[0021] In this specification, the spatial frequency or spatial frequency region may not necessarily be high or low as generally considered to be high or low, and is set based on a predetermined standard in the makeup support according to this embodiment. For example, the high spatial frequency region may be 60c / fw or higher, and the low spatial frequency region may be less than 60c / fw. The high spatial frequency region and the low spatial frequency region may be separated by a predetermined value, and in this case, the predetermined value may be 40c / fw or higher and 80c / fw or lower, for example, 60c / fw. In this specification, the unit of spatial frequency, c / fw (cycle / face width), is the number of cycles when the face width in the binarized image is used as the unit length.

[0022] The spectrum analysis can determine which makeup patterns show relatively large power spectrum values ​​(power spectral density) in the high spatial frequency region, i.e., contribute to an increase in the high spatial frequency components, and can also determine which makeup patterns show relatively large power spectrum values ​​in the low spatial frequency region, i.e., contribute to an increase in the low spatial frequency components.

[0023] In the above-mentioned spectral analysis, the spatial frequency characteristics of the entire acquired face image may be obtained, or the spatial frequency characteristics of the image of the part where makeup is applied may be obtained. For example, an image of a part where eye makeup is applied, including the eyes and eyebrows, or an image of the mouth and its surrounding area where lip makeup is applied, may be cut out to obtain a partial image, and the spatial frequency of the partial image may be subjected to spectral analysis. A more accurate database can be obtained by analyzing the part where makeup is applied.

[0024] Next, the amount of optical illusion is measured for each of the multiple face images for each different observation situation (S03). As described above, the observation situation of the face may be a specific situation such as whether it is a situation in which the face is viewed in real life or a situation in which the face is viewed on a display, and further, in the former case, the observation distance, etc., and in the latter case, the resolution of the display, etc. In addition, the observation situation may be classified as described above, that is, either a real close distance (observation situation I) or a real long distance or display (observation situation II). In other words, the amount of optical illusion (optical illusion effect) may be measured for the multiple face images under each of the conditions of observation situation I and observation situation II.

[0025] The amount of illusion is the amount of error from the actual state caused by a visual illusion. More specifically, the amount of illusion caused by makeup refers to the degree to which the size and / or shape of facial features perceived by an observer differs from the actual one. For example, in eye makeup, the amount of illusion is how much larger or smaller the eyes appear than the actual size, and in lip makeup, how much larger or smaller the mouth appears than the actual size, or how much the shape of the mouth differs from the actual size. For example, the staircase method (up and down method) commonly used in the field of psychology can be used to measure the amount of illusion. In this case, if the amount of illusion regarding the size of the eyes is to be measured, for example, a plurality of images in which the size of the eyes is changed to larger and smaller by image processing is created from a face image without makeup (eye size 100%), and such images can be used as comparative stimulus images. Then, under a certain observation situation or under a situation simulating the observation situation, one or more panelists are made to evaluate the amount of illusion (%) of each face image, which is the standard stimulus image. Such measurement of the amount of illusion is performed in the same manner for different observation situations.

[0026] By collecting the data obtained in the above manner in advance, registering it, and storing it in a database, it is possible to obtain a makeup pattern corresponding to a predetermined observation situation and a predetermined amount of optical illusion. In addition, a relationship between the spatial frequency characteristics of the face image or makeup pattern and the amount of optical illusion for each observation situation may be obtained from the obtained data (S04). The relationship may be, for example, a formula expressing the relationship between the feature amount in a predetermined spatial frequency domain and the amount of optical illusion. This makes it possible to propose a makeup pattern suitable for the observation situation and the amount of optical illusion (for example, the size of the eyes to be perceived) desired by the subject. In addition, it is possible to evaluate whether the makeup of the subject is suitable for the amount of optical illusion of the observation situation desired by the subject (for example, the size of the eyes to be perceived).

[0027] <How to suggest makeup> One embodiment of the present invention is a method for suggesting makeup, which includes acquiring information about the observed state of the face of a person to whom the makeup is to be applied (S11), and proposing makeup that provides predetermined spatial frequency characteristics suitable for the observed state based on the information (S12), as shown in Fig. 2. Each step of the makeup suggestion method according to this embodiment will be described in detail below, but the description is mainly based on an example of an eye makeup suggestion method.

[0028] In the makeup suggestion method according to this embodiment, first, information on the observed state of the face of a subject to whom makeup is to be applied is obtained (S11). The information on the observed state of the face can be obtained by having the subject input the observed state of the face that the subject desires into a makeup suggestion device (described later), or it can be obtained by an operator such as a beauty consultant asking the subject questions.

[0029] The information on the observed situation may be specific information such as whether the observed situation is a real situation or a situation seen on a display, and further, in the former case, the observed distance, etc., and in the latter case, the display resolution, etc. The information on the observed situation may also be the classification as described above, that is, information on whether a real close distance (observed situation I) or a real long distance or display display (observed situation II) has been selected. Furthermore, the operator or makeup suggestion device that has acquired specific information on the observed situation may classify it into either observed situation I or observed situation II.

[0030] Furthermore, a desired amount of optical illusion, i.e., information on how much the size and / or shape of facial features should look different from the actual size, can be obtained from the subject. In the case of eye makeup support, the desired amount of optical illusion may be a request regarding the size of the eyes as perceived by the observer. That is, the makeup suggestion method may further include a step of obtaining the subject's request regarding the size of the eyes as perceived by the observer. Such a request can be obtained, for example, by obtaining a face image of the subject without makeup, preparing a plurality of images in which the size of the eyes of the face image is changed to 100% and the size of the eyes is changed to different degrees, and having the subject select an image with the desired eye size from the plurality of images. In this case, the subject may be allowed to select from a plurality of stages, such as three stages of large, medium, and small. However, the amount of optical illusion may be set to an appropriate constant value that is generally preferred. In that case, the step of obtaining the desired amount of optical illusion from the subject (the step of obtaining the subject's request regarding the size of the eyes as perceived by the observer) can be omitted.

[0031] The makeup suggestion method according to the present embodiment proposes a makeup pattern that brings about a predetermined spatial frequency characteristic based on the subject's desired observation situation obtained as described above, or based on the subject's desired observation situation and the amount of optical illusion (S12). In this case, the above-mentioned database can be used to select a makeup pattern that corresponds to the subject's desired observation situation, or the subject's desired observation situation and the amount of optical illusion. The proposal may be presented, for example, in a form in which an image of a makeup pattern included in the database is displayed on a display and presented to the subject, or an explanation of the image is prepared and presented to the subject. Also, a no-makeup image of the subject may be obtained, and a pure makeup pattern image (an image obtained by removing the bare face portion from the makeup pattern image) may be superimposed on the image and displayed on the display. Alternatively, a beauty consultant or the like may view an image of a makeup pattern selected from the database and actually apply the makeup to the subject's face.

[0032] <Specific examples of makeup suggestions> Further, a specific example of the makeup suggestion method is shown in FIG. 3. In this method, first, information on the observed state of the face of the subject to whom makeup is applied (information on the desired observed state) is obtained (S31). In this example, the observed state can be selected from real close distance (observed state I) and real long distance or display (observed state II). Next, information on the desired perceived eye size (information on the desired amount of optical illusion) is obtained (S32a, S32b). In this example, the eye size is divided into three levels, large, medium, and small, and can be selected from among these. In this example, "large" is 105% or more of the actual eye size, "medium" is 100% to 104% of the actual eye size, and "small" is the same as or smaller than the actual eye size.

[0033] Next, the information on the subject's desired observation situation and the desired amount of illusion obtained in step 31, 32a or 32b (S31, S32a or S32b) is checked against the above-mentioned database constructed in advance to find a makeup pattern that brings about spatial frequency characteristics corresponding to the observation situation and the amount of illusion. As described above, for example, in observation situation I (real close distance), the illusion effect is large due to a makeup pattern that contains relatively many high spatial frequency components (makeup pattern that shows a relatively large power spectrum value in the high spatial frequency region), and in observation situation II (real long distance or display), the illusion effect is large due to a makeup pattern that contains relatively many low spatial frequency components (makeup pattern that shows a relatively large power spectrum value in the low spatial frequency region). Therefore, when trying to increase the illusion effect, makeup patterns (MP4, MP5) that contain relatively many high spatial frequency components can be proposed for observation situation I, and makeup patterns (MP1, MP2) that contain relatively many low spatial frequency components can be proposed for observation situation II.

[0034] FIG. 4 shows a makeup proposal example. Note that the expressions of the makeup patterns MP1 to MP6 shown in FIG. 4, both verbal and pictorial, are merely examples. In the observed situation I, the makeup pattern can be proposed so that the higher the perceived eye size, the more the high spatial frequency components are increased. For the observed situation I in FIG. 4, a makeup pattern with more high spatial frequency components can be proposed when the desired eye size is "large", and a makeup pattern with slightly more high spatial frequency components can be proposed when the desired eye size is "medium". For the observed situation II, a makeup pattern with more low spatial frequency components can be proposed when the desired eye size is "large", and a makeup pattern with slightly more low spatial frequency components can be proposed when the desired eye size is "medium". In both the observed situation I and the observed situation II, when the desired eye size is "small", a pattern with less makeup itself is proposed to suppress the illusion effect of making the eyes larger, but in the observed situation II, since the eyes are originally easily misperceived as being larger, it is preferable to propose the use of a concealer or the like with the intention of making the eyes look a little smaller.

[0035] <Makeup suggestion device> The suggestion method as described above can be performed by a makeup suggestion device. Thus, one embodiment of the present invention is a makeup suggestion device that includes an acquisition unit that acquires information regarding an observed state of a face of a subject to which makeup is to be applied, and a suggestion unit that suggests eye makeup suitable for the observed state based on the information, and the suggested eye makeup may include makeup that provides a predetermined spatial frequency characteristic.

[0036] Fig. 5 shows an example of a functional block diagram of a makeup suggestion device. As shown in Fig. 5, makeup suggestion device 10 may include an acquisition unit 11, an analysis unit 12, a registration unit 13, and a suggestion unit 14. By executing a program, makeup suggestion device 10 can function as acquisition unit 11, analysis unit 12, registration unit 13, and suggestion unit 14. Note that makeup suggestion device 10 can be used not only for the makeup suggestion method, but also for the above-mentioned database construction method.

[0037] The acquiring unit 11 acquires information on the subject's observed state in the above-mentioned step 11 (S11). The acquiring unit 11 can also acquire information related to makeup suggestions other than the information on the observed state, such as the amount of optical illusion of facial features desired by the subject (in the case of eye makeup, a request regarding the perceived size of the eyes).

[0038] Furthermore, the acquiring unit 11 can also acquire information (data) of the facial image of the target model when constructing the database (steps 01 to 04 above). The facial image information may be an uncovered facial image of the target model, or a facial image of the target model with makeup applied. The facial image information may be acquired by the acquiring unit 11 via an imaging device (not shown) connected to the makeup suggestion device 10. The acquiring unit 11 can also perform some kind of processing on the acquired facial image. This processing may include a process of superimposing an image of a makeup pattern on the uncovered facial image to synthesize it, and a process of generating an image of such a makeup pattern.

[0039] The analysis unit 12 analyzes the information obtained by the acquisition unit 11. For example, particularly when constructing a database (steps 01 to 04 above), the analysis unit 12 can perform a spectral analysis of the spatial frequency of the made-up image (including an image of a face with made-up applied and an image of a face without makeup superimposed with a make-up pattern). This spectral analysis may include image processing for the spectral analysis, such as grayscale processing, binarization processing, and filtering processing. In this case, the analysis unit 12 can also create a power spectrum diagram and extract a feature amount of a predetermined spatial frequency.

[0040] The registration unit 13 registers the acquired information, information obtained by processing or altering the acquired information, information input from the outside, etc. Such information may be, for example, information on the subject's observation situation acquired by the acquisition unit 11, face image information, etc., and information on the analysis results of the power spectrum diagram, feature amount, etc. created by the analysis unit 12.

[0041] In step 12 (S12), the suggestion unit 14 suggests makeup suitable for the observed situation based on information about the observed situation of the subject's face. Furthermore, when the amount of optical illusion for the subject's desired facial features (information about the perceived size of the eyes in the case of eye makeup) is acquired, appropriate makeup that meets the subject's desires is suggested based on the information about the desired amount of optical illusion and the information about the observed situation. The suggestion can utilize the database constructed through the above-mentioned steps 01 (S01) to 04 (S04). The makeup suggestions made by the suggestion unit 14 can be displayed via a display device (not shown) such as a display connected to the makeup suggestion device 10.

[0042] <Makeup evaluation method> One embodiment of the present invention is a method for evaluating makeup, which includes acquiring an image of a face of a subject to which makeup has been applied (S21), and evaluating the makeup for suitability to the observed state of the subject's face based on the spatial frequency characteristics of the made-up portion of the image (S22), as shown in Fig. 6. Each step of the makeup evaluation method according to this embodiment will be specifically described below, but the description is mainly based on an example of an eye makeup evaluation method.

[0043] The makeup evaluation method can also use the above-mentioned database (constructed in steps 01 to 04, FIG. 1). Using such a database, it is possible to determine what kind of observation situation the makeup applied to the subject is suitable for and / or what level of optical illusion it will produce. It is also possible to evaluate whether or to what extent the makeup applied to the subject is suitable for the observation situation and / or optical illusion level (e.g., the desired perceived eye size) desired by the subject.

[0044] In the makeup evaluation method, first, an image of the subject's face with makeup applied is acquired (S21), and the acquisition method is the same as the method in step 01 (S01) in constructing the database.

[0045] Then, when evaluating the makeup of the acquired face image of the subject (S22), the spatial frequency characteristics of the makeup image are used. That is, the spatial frequency characteristics are obtained for the image of the face of the subject with makeup, preferably for the part where makeup is applied. The method of obtaining the spatial frequency characteristics may be the same as the method performed in step 02 (S02) of obtaining the spatial frequency characteristics in the construction of the database. By obtaining the spatial frequency characteristics in the same format as the spatial frequency characteristics registered in the database, it becomes easy to check against the database. For example, if the relationship between the feature amount in a predetermined spatial frequency region and the amount of optical illusion is registered in the database, the same type of feature amount in the same predetermined spatial frequency region is obtained when obtaining the spatial frequency characteristics in the evaluation step (S22).

[0046] In the evaluation of makeup (S22), for example, the optical illusion effect of makeup (in the case of eye makeup, the effect of making the eyes look larger) is set to a suitable fixed value that is generally preferred, and an observation situation in which the spatial frequency characteristics of the facial image of the subject with makeup applied are deemed appropriate can be selected from the relationship (described above) between the spatial frequency characteristics of makeup in the database, the observation situation, and the amount of optical illusion. The observation situation may be shown in concrete terms such as a specific observation distance at which the face is actually seen and the resolution of the display on which the face is displayed, as described above, or may be shown in a rough classification of either a real close distance (situation I) and a real long distance or display display (situation II).

[0047] In addition, in the makeup evaluation method, the desired illusion effect or amount of illusion (in the case of eye makeup, the desire for the perceived size of the eyes) can be acquired from the subject. Then, the spatial frequency characteristics of the image of the face of the subject with makeup and the desired illusion effect or amount of illusion can be compared with the spatial frequency characteristics of makeup in the database, the observed situation, and the relationship (described above) between the observed situation and the amount of illusion to determine an appropriate observed situation. Furthermore, if the desired observed situation is acquired from the subject, it is possible to determine the difference between the observed situation deemed appropriate in the evaluation step (S22) and the observed situation desired by the subject. The format for presenting the results of the evaluation is not particularly limited, and may be presented via a display (display device) or by voice.

[0048] <Specific examples of makeup evaluation methods> Furthermore, an example of a specific makeup evaluation method is shown in Fig. 7. The method of this example includes obtaining an image of a subject's face without makeup (no-makeup image) (S41), obtaining an image of the same subject's face with makeup (made-up image) (S42), and performing power spectrum analysis of spatial frequencies on both the no-makeup image and the made-up image (S43).

[0049] In this example, when evaluating the made-up image, the analysis result of the no-makeup image is used as a standard. For this purpose, in the power spectrum analysis of the images (S43), the spatial frequency characteristics are obtained for the no-makeup image and the made-up image. More specifically, for the no-makeup image, a feature quantity Vn H and the feature quantity Vn L In addition, for makeup images, we also calculate the feature quantity Vm H and the feature quantity Vm LThe high spatial frequency region and the low spatial frequency region may have a predetermined spatial frequency band, but the high spatial frequency region and the low spatial frequency region may be divided by a predetermined spatial frequency. The boundary may be a value within a range of 40c / fw or more and 80c / fw or less, for example, 60c / fw. As described above, the feature amount may be one or more of a representative value such as an average value or a median value, and an integral value of the power spectrum value in the predetermined spatial frequency region.

[0050] Furthermore, in the method of this example, the feature amount Vm H and the high-frequency feature vector Vn H Compared with (Vm H -Vn H )>θ H It is determined whether or not (S44). H -Vn H )>θ H Even if (Vm H -Vn H )≦θ H Even if the feature amount Vm L and the low-frequency feature vector Vn L Compared with (Vm L -Vn L )>θ L By making these determinations, makeup can be classified into four types.

[0051] Then, by comparing the results with a database (Figure 1) that has been prepared in advance, specifically, by comparing the respective feature amounts in the high spatial frequency region and the low spatial frequency region obtained under the same conditions as in this example with the relationship between the observed situation and the observed situation, a diagnosis is made as to which observed situation each of the classified makeup types is suited.

[0052] FIG. 8 shows an example of the makeup diagnosis result. The expression formats of the makeup diagnosis results DN1 to DN4 are also merely examples. In this example, a face without makeup is used as a reference, so the feature amount Vm H and the low frequency region feature value Vm L If either of these values ​​is lower than that of the no-makeup image, it is diagnosed as makeup that does not have the expected optical illusion effect.

[0053] <Makeup evaluation device> The makeup evaluation method as described above can be performed by a makeup evaluation device. Thus, one embodiment of the present invention may be a makeup evaluation device that includes an acquisition unit that acquires an image of a face of a subject wearing makeup, an analysis unit that performs an analysis to determine spatial frequency characteristics of the makeup part of the image, and an evaluation unit that evaluates the makeup for suitability to the observed state of the subject's face based on the spatial frequency characteristics.

[0054] Fig. 9 shows an example of a functional block diagram of a makeup evaluation device. As shown in Fig. 9, makeup evaluation device 20 may include an acquisition unit 21, an analysis unit 22, a registration unit 23, and an evaluation unit 24. By executing a program, makeup evaluation device 20 can function as acquisition unit 21, analysis unit 22, registration unit 23, and evaluation unit 24. Note that makeup evaluation device 20 can be used not only for the makeup evaluation method, but also for the above-mentioned database construction method.

[0055] The acquiring unit 21 acquires an image of the subject's face with makeup applied in the above-mentioned step 21 (S11). In addition to acquiring images, the acquiring unit 11 can also acquire information related to makeup evaluation, such as information about the subject's desired observation situation, such as the subject's desired amount of optical illusion for facial features (in the case of eye makeup, a desire for the perceived size of the eyes), etc.

[0056] The facial image information can be acquired by the acquisition unit 21 via an imaging device (not shown) connected to the makeup evaluation device 20. The acquisition unit 21 can also perform some processing on the acquired facial image. Furthermore, the acquisition unit 21 can also acquire facial image information (data) of the target model when constructing the database (steps 01 to 04 above). The facial image information when constructing the database may be an uncovered facial image of the target model, or an image of the target model with makeup applied. The acquisition unit 21 can also perform a process of superimposing an image of a makeup pattern on an uncovered facial image to synthesize it, and a process of generating an image of such a makeup pattern.

[0057] The analysis unit 22 analyzes the information obtained by the acquisition unit 21. For example, the analysis unit 22 can perform a spectrum analysis of the spatial frequency of the facial image of the subject with makeup applied. This spectrum analysis may include image processing for the spectrum analysis, such as grayscale processing, binarization processing, and filtering processing. The analysis unit 22 can also create a power spectrum diagram and extract features of a predetermined spatial frequency. The analysis unit 22 can also analyze a makeup image (including an image of a face with makeup applied and an image of a natural face image superimposed with a makeup pattern) particularly when constructing a database (steps 01 to 04 above).

[0058] The registration unit 23 registers the acquired information, information obtained by processing or altering the acquired information, information input from the outside, etc. Such information may be, for example, information on the image of the face of the subject to be evaluated acquired by the acquisition unit 21, information on the amount of optical illusion desired by the subject in some cases, and information on the observation situation desired by the subject, as well as information on the analysis results such as a power spectrum diagram and feature amounts created by the analysis unit 22.

[0059] In step 22 (S22), the evaluation unit 24 evaluates the makeup for suitability to the observed situation of the subject's face based on the spatial frequency characteristics of the made-up portion of the facial image of the subject. At that time, the evaluation unit 24 can compare the makeup with the database constructed through the above-mentioned steps 01 (S01) to 04 (S04) and present a corresponding suitable observed situation. In addition, if the subject has acquired the amount of optical illusion of the facial part desired by the subject (information on the perceived size of the eyes in the case of eye makeup), the evaluation unit 24 can present an observed situation suitable for the subject's makeup, taking into consideration the information on the desired amount of optical illusion. The evaluation result of the makeup by the evaluation unit 24 can be presented to the subject via a display device (not shown) such as a display connected to the makeup evaluation device 20.

[0060] Although the makeup suggestion method and makeup suggestion device, and the makeup evaluation method and makeup evaluation device have been described above, the makeup suggestion method and makeup evaluation method can be combined. Similarly, the makeup suggestion device and makeup evaluation device can be combined.

[0061] Therefore, in one embodiment of the present invention, the makeup proposed by the makeup suggestion method can be applied to a subject, and then the applied makeup can be evaluated by the makeup evaluation method. This embodiment may be a makeup support method including: acquiring information on the observed state of the face of the subject to be applied with the makeup; proposing makeup that brings about a predetermined spatial frequency characteristic suitable for the observed state based on the information; applying the proposed makeup to the subject; acquiring an image of the face of the subject to which the makeup has been applied; and evaluating the suitability of the eye makeup with the observed state of the face of the subject based on the spatial frequency characteristic of the part of the image to which the eye makeup has been applied. The makeup support method may further include correcting or redoing the makeup based on the evaluation. After correcting or redoing the makeup, the makeup may be evaluated again, and after the evaluation, the correction or redoing of the makeup and the makeup evaluation may be repeated. Such a makeup support method can avoid a situation in which the proposed makeup is actually applied but a desired optical illusion effect cannot be obtained in a predetermined observed state.

[0062] <Hardware configuration> Fig. 10 is a block diagram showing an example of the hardware configuration of the above-mentioned makeup suggestion device 10 or makeup evaluation device 20. The hardware configuration shown in Fig. 10 may also be a block diagram of a makeup support device in which the makeup suggestion device and the makeup evaluation device are combined. Note that each element in the hardware configuration is connected to each other via a bus B.

[0063] The makeup suggestion device 10 and / or the makeup evaluation device 20 may include a central processing unit (CPU) 1001, a read only memory (ROM) 1002, and a random access memory (RAM) 1003. The CPU 1001, the ROM 1002, and the RAM 1003 form a so-called computer.

[0064] The makeup suggestion device 10 and / or the makeup evaluation device 20 may include an auxiliary storage device 1004 , a display device 1005 , an operation device 1006 , an I / F (Interface) device 1007 , and a drive device 1008 .

[0065] The CPU 1001 is a computing device that executes various programs installed in the auxiliary storage device 1004 .

[0066] The ROM 1002 is a non-volatile memory. The ROM 1002 functions as a main storage device that stores various programs, data, etc. required for the CPU 1001 to execute various programs installed in the auxiliary storage device 1004. Specifically, the ROM 1002 functions as a main storage device that stores boot programs such as a BIOS (Basic Input / Output System) and an EFI (Extensible Firmware Interface).

[0067] The RAM 1003 is a volatile memory such as a dynamic random access memory (DRAM) or a static random access memory (SRAM). The RAM 1003 functions as a main storage device that provides a working area in which various programs installed in the auxiliary storage device 1004 are expanded when the CPU 1001 executes them.

[0068] The auxiliary storage device 1004 is an auxiliary storage device that stores various programs and information used when the various programs are executed.

[0069] The display device 1005 is a display device that displays the contents of information acquired by each of the makeup suggestion device 10 and / or the makeup evaluation device 20, the contents of results output by the devices, and the like.

[0070] The operation device 1006 is an input device through which a user of the device inputs various instructions to the makeup suggestion device 10 and / or the makeup evaluation device 20.

[0071] The I / F device 1007 is a communication device that connects to a network and communicates with the makeup suggestion device 10 and / or the makeup evaluation device 20.

[0072] The drive device 1008 is a device for setting the storage medium 1009. The storage medium 1009 here includes media that record information optically, electrically, or magnetically, such as a CD-ROM, a flexible disk, a magneto-optical disk, etc. The storage medium 1009 may also include semiconductor memories that record information electrically, such as an EPROM (Erasable Programmable Read Only Memory), a flash memory, etc.

[0073] The various programs to be installed in the auxiliary storage device 1004 are installed, for example, by setting the distributed storage medium 1009 in the drive device 1008 and reading out the various programs recorded in the storage medium 1009 by the drive device 1008. Alternatively, the various programs to be installed in the auxiliary storage device 1004 may be installed by being downloaded from a network via the I / F device 1007. EXAMPLES

[0074] An experimental example on eye makeup is shown below as an example. In this experiment, an image IM0 (no makeup image) of a face of a target model (subject) without makeup, and images IM1 and IM2 (makeup images) of the face of the same target model with eye makeup were created. Images IM1 and IM2 (makeup images) were created by processing image IM0 (no makeup image) using the layer function of Photoshop (registered trademark) (Adobe). Specifically, a makeup pattern layer was created to be superimposed on the area including the eyebrows and eyes of the target model in image IM0, and the layer was superimposed on image IM0 to obtain images IM1 and IM2. At that time, a makeup pattern with a thickness generally recognized as very heavy makeup was created with an opacity of 100%, and the image IM1 was superimposed on image IM0. Furthermore, a layer with the same opacity as the above layer changed to 67% was created, and the image IM2 was superimposed on image IM0. The makeup pattern in the created layer was a pattern that represented makeup using eyeshadow, eyeliner (above and below the eyes), and mascara, with the eyeshadow having a gradation that gradually became darker from the top to the bottom of the eye socket.

[0075] Next, an area including eyebrows and eyes was cut out from each image (the cut out area is shown by a box in FIG. 11), and the obtained partial image was grayscale processed using MATLAB (registered trademark) (manufactured by MathWorks). Furthermore, the power spectrum in the vertical direction (horizontal stripes) was calculated by two-dimensional Fourier transform using MATLAB (registered trademark) (manufactured by MathWorks). FIG. 12 shows the power spectrum diagram of the partial images of image IM0 (no makeup image) and images IM1 and IM2 (made-up images). FIG. 12 shows that the spatial frequency characteristics are different depending on the makeup, that is, different makeup (images IM1 and IM2) bring about different spatial frequency characteristics. Furthermore, when the power spectrum values ​​of images IM1 and IM2 were compared, the power spectrum value of image IM1 was higher than that of image IM0 in both the high spatial frequency region (e.g., 85c / fw) and the low spatial frequency region (e.g., 35c / fw). In other words, image IM1 had a higher power spectrum value in the entire frequency region. On the other hand, image IM2 has a high power spectrum value in the high spatial frequency region (eg, 85c / fw), but its power spectrum value in the low spatial frequency region (eg, 35c / fw) is almost the same as that of image IM0.

[0076] On the other hand, we quantified how much larger the target model's eyes appear to be than their actual size when observers viewed images IM1 and IM2 from different distances. Specifically, 32 panelists were shown images (IM1 and IM2) created with different opacity layers on a display, and the panelists were asked to answer how large the target model's eyes appeared to be by changing the position of the display to change the distance between the panelists and the display. The degree of magnification when displaying the images was set to two types, corresponding to the distance (observed distance) between the panelists and the target model when they actually viewed it being 0.6 m and 5 m. The panelists used the staircase method (up and down method) commonly used in the field of psychology for their evaluation.

[0077] In preparation for the evaluation by the staircase method (up and down method), the size of the target model's eyes in image IM0 (no makeup image) was set as 100%, and the size of the target model's eyes was changed from 90% to 110%, and images (comparison stimulus images) were obtained for every 2% change. In other words, 11 images with different eye sizes of 90%, 92%, ..., 100%, ..., 108%, and 110% were created. Then, the target image (image IM1, IM2, or IM3) was displayed as a standard stimulus image on the display for the panelists, and the comparison stimulus images with different eye sizes were presented next to the standard stimulus image. The panelists were asked to choose which image had the target model with larger eyes. If the panelists chose the standard stimulus image, the comparison stimulus image was changed to an image with larger eyes in the next trial, and if the panelists chose the comparison stimulus image, the comparison stimulus image was changed to an image with smaller eyes in the next trial, and the panelists were asked to choose the image with the larger eyes again. When changing the image, the eye size was changed to an image that was 2 points larger or smaller, and the panelists continued to select the image until the selection responses between the standard stimulus and the comparison stimulus were balanced. The eye size (%) at which the panelists' answers were balanced was determined and recorded as the amount of optical illusion (%). Therefore, the amount of optical illusion (%) is a value that represents how much the eyes of the target model in each image appeared larger or smaller than the actual eye size without makeup to the panelists. The measurements by all panelists were arithmetically averaged, and the results are shown in Figure 13. For comparison, the amount of optical illusion (%) was also measured for image IM0 in the same way.

[0078] As shown in Fig. 13, the magnitude of the illusion in both images IM1 and IM2 was greater at longer observed distances. Also, according to the graph in Fig. 13, in this experimental example, the slope was greater for image IM1 than for image IM2, and the greater the distance, the greater the effect of image IM1 (makeup with more high and low spatial frequency components, Fig. 12) in increasing the magnitude of the illusion. Also, at close distances, there was a tendency for the magnitude of the illusion in images IM1 and IM2 (makeup with more high spatial frequency components only, Fig. 12) to be about the same. This is thought to be because, at close distances, the difference in the spectral value in the high frequency range with image IM0 affects the illusion effect, but because both images IM1 and IM2 have higher spectral values ​​in the high frequency range than image IM0, the illusion effect was seen at close distances for both.

[0079] In this way, it was found that the makeup that is effective in increasing the amount of the optical illusion (in this experimental example, the effect of making the eyes appear larger) differs depending on the observation distance. Based on this knowledge, it is possible to suggest makeup to the user according to the observation distance, and also to determine whether the user's makeup is suitable for a given observation distance.

[0080] Although the present invention has been described above based on the embodiments, the present invention is not limited to these embodiments. Furthermore, the above embodiments can be changed, modified, substituted, added, deleted, and combined in various ways within the scope of the claims, and these also belong to the technical scope of the present invention. [Explanation of symbols]

[0081] 10 Makeup suggestion device 11 Acquisition Department 12 Analysis Department 13. Registration Department 14 Proposal Department 20 Makeup evaluation device 21 Acquisition Department 22 Analysis Department 23 Registration Department 24 Evaluation Section 1001 CPU 1002 ROM 1003 RAM 1004 Auxiliary storage device 1005 Display device 1006 Operating device 1007 I / F device 1008 Drive device 1009 Storage medium

Claims

1. obtaining information regarding the observation status of the face of the subject to whom eye makeup is to be applied; Proposing eye makeup suitable for the observed situation based on the information, The makeup suggestion method, wherein the suggested eye makeup includes makeup that results in a predetermined spatial frequency characteristic.

2. The method of claim 1 , wherein the spatial frequency characteristics include a ratio of high spatial frequency components in a spatial frequency power spectrum and / or a ratio of low spatial frequency components in the spatial frequency power spectrum.

3. The makeup suggestion method according to claim 1 or 2, wherein the suggested eye makeup is one or more selected from the group including a type of makeup product, and an application position, an application range, and a gradation gradient of the makeup.

4. 3. The makeup suggestion method according to claim 1 or 2, wherein the information regarding the observed situation includes information regarding the distance of observation when the face of the subject to whom eye makeup is to be applied is observed directly, and includes information regarding processing of the image when the face of the user to whom eye makeup is to be applied is observed through an image.

5. Further obtaining the subject's desire regarding the size of the eyes as perceived upon the observation; 3. The makeup suggestion method according to claim 1, wherein the suggested eye makeup is eye makeup that is also suited to the request.

6. Obtaining an image of the subject's face with eye makeup applied; A makeup evaluation method for evaluating the suitability of the eye makeup with the observed state of the subject's face based on the spatial frequency characteristics of the made-up portion of the image.

7. 7. The makeup evaluation method according to claim 6, wherein the spatial frequency characteristics include a proportion of high spatial frequency components in a spatial frequency power spectrum for the eye makeup portion of the image, and / or a proportion of low spatial frequency components in the spatial frequency power spectrum.

8. In order to obtain the spatial frequency characteristics, performing a spatial frequency power spectrum analysis on the eye makeup portion of the image; The method further includes calculating a feature amount related to the intensity of a spatial frequency component in a predetermined spatial frequency band; The makeup evaluation method according to claim 6 or 7, wherein the evaluating comprises comparing the feature amount with a threshold value, and evaluating the eye makeup based on a result of the comparison.

9. 8. The makeup evaluation method according to claim 6 or 7, wherein the information regarding the observation situation includes information about the observation distance when the face of the subject to which eye makeup is applied is observed directly, and includes information about processing of the image when the face of the user to which eye makeup is applied is observed through an image.

10. Further obtaining the subject's desire regarding the size of the eyes as perceived upon the observation; The makeup evaluation method according to claim 6 or 7, wherein the evaluation of the eye makeup also includes evaluation of conformity with the desire.

11. Applying makeup suggested by the makeup suggestion method according to claim 1 or 2, A makeup assist method, comprising: evaluating the applied eye makeup by the makeup evaluation method according to claim 6 or 7.

12. An acquisition unit that acquires information regarding an observation state of a face of a subject to be applied with eye makeup; a suggestion unit that suggests eye makeup suitable for the observed situation based on the information, The proposed eye makeup includes makeup that provides a predetermined spatial frequency characteristic. , makeup suggestion device.

13. An acquisition unit for acquiring an image of a face of a subject having eye makeup applied thereto; an analysis unit that performs an analysis to obtain a spatial frequency characteristic of the eye makeup portion of the image; and an evaluation unit that evaluates the eye makeup for suitability to an observed state of the subject's face based on the spatial frequency characteristics.

Images