Cosmetic product color classification method and tool for assisting color development of cosmetic product

Abstract

A computer-implemented tool intended to be controlled by a user for assisting the development of cosmetic product colors, the tool comprising: a mapping mode adapted for the user to select a set of cosmetic product colors from a library of cosmetic product colors, the drawing mode is configured to classify each of the selected set of colors using a method of classifying the colors of the makeup product according to labels designed to subjectively perceive the makeup colors. The drawing mode is configured to display a graph of selected colors arranged in a table according to the respectively assigned labels, the table being composed in a main row representing each color system label and a main column representing each hue sub-system label, each main row including a sub-row representing each luminance sub-system label, each sub-row including a sub-row representing each luminance sub-system label, and each sub-column including a sub-row representing each luminance sub-system label. Each primary column includes a sub-column representing each chroma sub-system tag.

Description

Technical Field

[0001] Embodiments of the present invention relate to a method for classifying the colors of beauty and cosmetic products, the method being automatically implemented by a computer, and a tool for assisting in the development of the colors of beauty and cosmetic products, the tool being implemented by a computer and controlled by a user. Background Art

[0002] "Cosmetic product" means any product as defined in Regulation (EC) No 1223 / 2009 of the European Parliament and of the Council of 30 November 2009 on cosmetic products. Cosmetic cosmetic products or "cosmetic products" are more particularly intended to cover the surface of the body to modify perceived colour and / or texture.

[0003] Cosmetic product color development may include visualization and comparison of existing cosmetic product colors for different skin tones in a given market, for example, to identify trends in consumer preferred shades, or to identify shades that may be missing and needed in a given market.

[0004] For example, the market studied can range from low-cost products to luxury goods, and products can vary depending on the market location, for example, the Asian market is different from the European market.

[0005] Comparisons between cosmetic products can be made between different brands and franchise products of the same company, or between owned traditional products and competitors' products.

[0006] Visualization is usually done by color experts through physical samples of the color as perceived by the human eye (e.g., through thumbnail color samples of makeup products), because slight changes in absolute color can have a large impact on the subjective perception of makeup color. In fact, taking the red color of lipstick as an example, the first red may appear subjectively cold and lifeless, while the second red with an absolute value very close to the first red may appear subjectively warm and rich.

[0007] The absolute difference between two colors can be, for example, the distance that separates these colors in a "standard observer" model, such as the "CIEXYZ" color space or the "CIELAB" color space. These standard observer models are defined by the International Commission on Illumination (abbreviated to CIE), and the colors they define are not related to any specific device (such as a computer monitor or printer), but rather to the CIE standard observer, which is the average of the results of color matching experiments under laboratory conditions.

[0008] The RGB "Red Green Blue" color space is defined by the coordinates of the additive primary colors red, green and blue, and is commonly used in electronic devices to sense and display color.

[0009] The "CIELAB" color space, also known as L*a*b*, represents color as three coordinates: L* represents lightness, and a* and b* represent greenish to reddish chromaticity and bluish to yellowish chromaticity, respectively (chromaticity expresses both hue and chromaticity measures). In the a*b* plane, unique hues can be identified by unique angles in a trigonometric circle. The CIELAB color space is designed to be more perceptually uniform than, for example, the RGB color space. However, the subjective perception of the color of a cosmetic product is not well represented in the CIELAB color space, so that slight changes in absolute color in the CIELAB color space can still have a large effect on the subjective perception of the cosmetic color.

[0010] The human eye's perception of color goes beyond L*a*b* coordinates and values ​​subjective perceptions, such as the "background effect" of color, where, for example, viewing a single patch of color surrounded by a colored background results in a different perception of the patch depending on the color of the background. With regard to makeup products, the perception of a lipstick color may vary depending on the skin tone of the wearer.

[0011] Therefore, it is difficult to provide a computerized tool to assist color experts in developing the colors of cosmetic products because the digital transcription of color (such as RGB or L*a*b* coordinates) does not correspond to the subjective perception of cosmetic color.

[0012] Another difficulty encountered by computerized tools used to assist color experts in developing colors for cosmetic products is that the range of colors displayed by computer monitors varies, and all monitors have a limited range of colors (formerly known as "color number"). The resulting problem is that two different colors that have distinguishable properties under human perception may appear exactly the same on various monitors and therefore cannot be used for analysis by cosmetic color experts.

[0013] Therefore, visualization and comparison of cosmetic product colors is still typically performed under laboratory conditions via thumbnail color samples of cosmetic products (usually in very large quantities), and cannot be performed remotely via a computer (e.g., under home office conditions).

[0014] Another difficulty in developing makeup product colors is that the applied color of the product may vary depending on the skin tone of the makeup user, and the subjective perception of color may also vary depending on the skin tone of the user. For example, as mentioned earlier, the "background effect" of skin tone can change the perception of the color of a makeup product, and essentially, a light or pale makeup color may appear darker when used on dark skin, and a dark and rich makeup color may appear too intense when used on fair skin while being a perfect match on dark skin.

[0015] Therefore, the development of makeup product colors usually requires a trial phase on mannequins to analyze the actual subjective perception of makeup when applied. These trial phases are usually time-consuming and expensive.

[0016] Therefore, there is a need to provide a tool, particularly a computerized tool, to assist color experts in developing the colors of cosmetic products and save time, which tool will adjust based on the subjective perception of cosmetic color, will not be limited by computer monitor performance (such as "color number"), and will provide the ability to develop cosmetic colors suitable for a variety of skin tones. Summary of the invention

[0017] According to one aspect of the present invention, a computer-implemented method for automatically classifying the colors of cosmetic products comprises:

[0018] - Provide input coordinate data of cosmetic product color in L*a*b* space;

[0019] - assigning a color family label (e.g., brown, pink, orange, purple, red) to a cosmetic product color based on identifying a color family volume in a set of color family volumes in L*a*b* space that contains the input coordinate data;

[0020] These color palettes are designed based on the subjective perception of makeup color, and makeup product colors are categorized according to the assigned labels.

[0021] The subjective perception of makeup color used for volume design and the boundary values, lines and surfaces defined below are advantageously defined by a human color expert, for example based on the aforementioned sensitive subjective perception specific to makeup color. Advantageously, the human color expert can provide a visual database of discrete color points, and the design of volume and boundary values, lines and surfaces can be performed by a data-driven calculation configured to fit the visual database to obtain continuity of color points. For example, the data-driven calculation configured to fit the visual database can be implemented by a model trained by machine learning.

[0022] In other words, the computer-implemented method according to this aspect allows for the classification of colors based on labels representing subjective perceptions of makeup colors, which labels are defined by conditions established by subjective perceptions specific to makeup colors. Thus, despite the limitations of computer representations of colors (color spaces) and on-screen color displays, the resulting classification (e.g., applied to each color of a set of analyzed makeup products) can provide useful information to color experts conducting makeup product color development.

[0023] According to an embodiment, the set of color system capacities in the L*a*b* space is generated from a database including a finite number of points in the L*a*b* space and by mathematical calculations, each point being labeled with a corresponding system label based on a subjective perception of the makeup color, the mathematical calculations comprising generating a triangulation of envelopes enclosing all points of each corresponding system and an interpolation of extending these envelopes until corresponding relative surfaces of adjacent envelopes match each other, the envelopes defining the enclosing circles of the corresponding color system capacities.

[0024] According to an embodiment, the color system capacity is configured to define brown, pink, orange, purple and red in the L*a*b* space according to the subjective perception of makeup color.

[0025] According to an embodiment, the method further comprises:

[0026] - assigning a brightness sub-label (e.g., light, medium, dark) to a cosmetic product color based on identification of the position of the input coordinate data relative to at least one brightness boundary value in a brightness coordinate axis L* of an L*a*b* space;

[0027] The at least one brightness boundary value is designed based on subjective perception of makeup color, and the makeup product color is classified according to the assigned label.

[0028] According to an embodiment, for input coordinates having a chromaticity value greater than a threshold set according to subjective perception of makeup color, the at least one brightness boundary value in the brightness coordinate axis L* of the L*a*b* space is reduced by one step.

[0029] According to an embodiment, the luma subsystem includes a bright label assigned when the luma input coordinate is greater than a first luma boundary value, a medium label assigned when the luma input coordinate is between the first luma boundary value and a second luma boundary value, and a dark label assigned when the luma input coordinate is below the second luma boundary value.

[0030] According to an embodiment, the method further comprises:

[0031] - assigning a chromaticity subsystem label (e.g., high, middle, low) to a cosmetic product color based on the position of the input coordinate data relative to at least one chromaticity boundary line in the a*b* plane of the L*a*b* space;

[0032] The at least one chromaticity boundary line is designed based on subjective perception of makeup color, and the makeup product color is classified according to the assigned label.

[0033] According to an embodiment, the at least one chromaticity boundary line in the a*b* plane of the L*a*b* space varies according to the hue of the input coordinate data based on a subjective perception of the makeup color.

[0034] According to an embodiment, the chromaticity subsystem includes a high label assigned when the chromaticity of the input coordinates is greater than a first chromaticity boundary line, an intermediate label assigned when the chromaticity of the input coordinates is between the first chromaticity boundary line and the second chromaticity boundary line, and a low label assigned when the chromaticity of the input coordinates is lower than the second chromaticity boundary line.

[0035] According to an embodiment, the method further comprises:

[0036] - assigning a hue subfamily label (e.g., cool, neutral, warm) to a cosmetic product color based on the position of the input coordinate data relative to at least one hue boundary surface within the corresponding hue volume in L*a*b* space;

[0037] The at least one hue boundary surface is designed based on a subjective perception of makeup colors, and the makeup product colors are categorized based on the assigned labels.

[0038] According to an embodiment, the at least one hue boundary surface in each color system volume of the L*a*b* space is defined according to a subjective perception of a makeup color.

[0039] According to an embodiment, the hue subset includes a warm label assigned when the input coordinates are located on one side of a first hue boundary surface, a neutral label assigned when the input coordinates are located between the other side of the first hue boundary surface and one side of a second hue boundary surface, and a cold label assigned when the input coordinates are located on the other side of the second hue boundary surface.

[0040] According to an embodiment, cosmetic product colors are hierarchically classified first according to color family labels, second according to hue sub-family labels, and then according to lightness sub-family labels and chroma sub-family labels.

[0041] This hierarchy allows for convenient classification of the displayed labels, e.g. for analyzing the color of cosmetic products. That is, other hierarchies may also be used, and regarding the process of assigning labels, the corresponding processing stages may be performed simultaneously, or perhaps the brightness label should be processed after the chroma label, since the brightness boundary value may depend on the chroma label.

[0042] According to another aspect, a computer-implemented tool, intended to be controlled by a user, for aiding in cosmetic product color development, the tool comprising:

[0043] - A drawing mode, which is suitable for a user to select a set of cosmetic product colors from a cosmetic product color library, is configured to classify each color in the selected set of colors using the method for classifying colors of cosmetic products as defined above, and is configured to display a graph of the selected colors arranged in a table according to the respectively assigned labels, the table being organized by a main row representing each color system label and a main column representing each hue sub-system label, each main row including a sub-row representing each lightness sub-system label, and each main column including a sub-column representing each chromaticity sub-system label.

[0044] According to an embodiment, the table is organized as follows:

[0045] - The five main rows represent the brown, pink, orange, purple and red labels respectively.

[0046] - The three main columns represent the cool, neutral, and warm sub-family labels,

[0047] - the three sub-rows of each main row represent the light, medium, and dark brightness sub-line labels, respectively, and

[0048] - The three sub-columns of each main column represent the low, middle and high chroma sub-line labels respectively.

[0049] According to an embodiment, the tool additionally comprises:

[0050] - Application mode, which is suitable for the user to select a set of at least one makeup product color from the displayed chart (ID card) and select at least one skin color photo model, and is configured to display simulated images of applying the selected set of makeup product colors on the at least one skin color photo model.

[0051] According to an embodiment, the tool additionally comprises:

[0052] - A color creation mode, which is suitable for a user to select at least one skin color photo model and set parameters for generating a custom color, and is configured to display a simulated image with the custom color applied on the at least one skin color photo model.

[0053] According to an embodiment, the color creation mode is further adapted for the user to select a set of at least one cosmetic product color from the displayed chart (ID card), and is configured to simultaneously display simulated comparison images of applying the custom color and the selected set of cosmetic product colors, respectively, on the at least one skin color photo model.

[0054] According to another aspect, a computer program product comprises instructions which, when the program is executed by a computer, cause the computer to perform a method as defined above, or cause the computer to perform a tool as defined above.

[0055] According to another aspect, a computer-readable storage medium comprises instructions which, when executed by a computer, cause the computer to perform a method as defined above, or cause the computer to perform a tool as defined above.

[0056] In other words, various aspects and embodiments provide, among other things, a digital system that enables instant analysis of color cosmetic shades and creation of shade ranges based on the following combined process: 1) data visualization of instrument measurement data based on labels suitable for color classification of cosmetic products; 2) digital application of measured shades on images depicting models of different skin tones, such as those taken using a pseudo-spectral imaging system; 3) the ability to digitally accurately conceive and apply shades on images, such as those taken using the aforementioned pseudo-spectral camera system.

[0057] Therefore, a method and system (tool) are defined that can integrate the development of shade ranges, particularly taking into account the evaluation of makeup on different skin tones, for use by makeup product development teams using a digital platform.

[0058] This enables product development teams to easily create shade ranges by providing the comprehensive ability to visualize shades from a database of measured shades and digitally conceive new shades, thereby creating better performing, skin tone-appropriate color cosmetic shade ranges.

[0059] Additionally, this improves the efficiency of tonal range development, potentially reducing tonal range development time by approximately 25% due to the ability to digitally visualize and compare lipstick shades in dedicated assortments and on-surface simulations.

[0060] Furthermore, since the aspects and embodiments defined above provide the ability to develop a range of shades deemed suitable for various skin tones, this may reduce the number of shades created each year, as the development team may visualize and select shades from a database to match a specific skin tone. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Other advantages and illustrations of the present invention will emerge when reviewing the non-limiting detailed description of embodiments in conjunction with the accompanying drawings, in which:

[0062] [ Figure 1 ];

[0063] [ Figure 2 ];

[0064] [ Figure 3 ];

[0065] [ Figure 4 ];

[0066] [ Figure 5 ];

[0067] [ Figure 6];

[0068] [ Figure 7 ] presents an embodiment of a method for classifying the colors of cosmetic products according to the present invention;

[0069] [ Figure 8 ];

[0070] [ Fig. 9 ] demonstrates the difficulties encountered in traditional techniques;

[0071] [ Fig.10 ];

[0072] [ Fig.11 ];

[0073] [ Fig.12 ];

[0074] [ Fig.13 ];

[0075] [ Fig.14 ] shows an embodiment of a tool for assisting in the color development of a cosmetic product according to the present invention. DETAILED DESCRIPTION

[0076] Figure 1 A method 100 for classifying the colors of a cosmetic product 102 is shown, the method being designed to be automatically implemented by a computer. The cosmetic product 102 is preferably a lipstick product. Here, a method for providing a classification label for a single color 104 of a single cosmetic product 102 is described. For example, at step 102, the cosmetic product 102 is selected from a cosmetic product color database. However, it is understood that the classification is primarily intended to be applied to several colors of several cosmetic products in order to distinguish these colors based on human subjective perception of cosmetic colors. In this case, the method is performed for each color of the group of several cosmetic products.

[0077] The method includes providing input coordinate data of the cosmetic product color 102 in L*a*b* space at step 104. As described above and in conjunction with the following Figure 2 As described, the L*a*b* space is a traditional "CIELAB" color space that represents color as three coordinates: L* represents lightness, and a* and b* represent greenish to reddish chromaticity and bluish to yellowish chromaticity, respectively (chromaticity expresses both hue and chromaticity metrics). Step 104 may include a traditional conversion of color coordinates from any other color space to coordinates in the L*a*b* color space.

[0078] The first assignment step 106 assigns a color family label to the cosmetic product colors 102 to 104 (for example, it may be a brown, pink, orange, purple, or red family). Figure 3As disclosed, a color system label can be assigned based on the identification of a color system capacity in a set of color system capacities in L*a*b* space that contains input coordinate data. In particular, the color system capacity is specifically designed based on the subjective perception of makeup color. For example and as described below in conjunction with Figure 3 As described, the color capacity is advantageously determined based on data driven calculations by fitting with a visual database.

[0079] The second assignment step 108 assigns the cosmetic product colors 102 to 104 a lightness subsystem label (e.g., it may be a light, medium, or dark lightness subsystem), a chroma subsystem label (e.g., it may be a high, middle, or low chroma subsystem), and a hue subsystem label (e.g., it may be a cool, neutral, or warm hue subsystem).

[0080] As combined with the following Figure 4 As described above, the brightness sub-system label can be assigned based on the identification of the position of the input coordinate data 104 relative to at least one brightness boundary value in the brightness coordinate axis L* of the L*a*b* space. In particular, the at least one brightness boundary value is specially designed based on the subjective perception of the makeup color.

[0081] As combined with the following Figure 5 As described above, the chromaticity subsystem label can be assigned according to the position of the input coordinate data 104 relative to at least one chromaticity boundary line in the a*b* plane of the L*a*b* space. In particular, the at least one chromaticity boundary line is specially designed according to the subjective perception of the makeup color.

[0082] As combined with the following Figure 6 As described above, a hue subsystem label can be assigned based on the position of the input coordinate data 104 relative to at least one hue boundary surface within the corresponding hue volume 106 in the L*a*b* space. In particular, the at least one hue boundary surface is specifically designed based on subjective perception of makeup color.

[0083] At the final step 110, the cosmetic product colors are categorized according to the assigned labels, for example and advantageously in an identification card diagram, as described below in conjunction with Figure 7 The layered arrangement.

[0084] Figure 2 An example discrete representation of the L*a*b* color space (also known as the CIELAB color space) is shown. The CIELAB color space represents color as three coordinates: L* represents lightness, and a* and b* represent green to red hues and blue to yellow hues, respectively.

[0085] L*a*b* space allows to easily obtain the hue, chromaticity and value of a color point in the L*a*b* coordinate system. The hue of a color point is identified by the triangular angle θ in the a*b* plane including the point, and can therefore be expressed as θ=tan -1 The chromaticity of a color point is defined by the distance from the origin (a*=0; b*=0) in the a*b* plane including the point, and can therefore be expressed as C*=(a*2+b*2) 1 / 2 The value of a color point is defined by its lightness coordinate L*.

[0086] Figure 3 The generation of the set of color system capacities in L*a*b* space is shown, and these color system capacities are used to identify the color system labels of the input coordinate data at the first assignment step 106 of the method 100 .

[0087] First, a database is provided that includes a limited number of points in L*a*b* space for each color family, wherein each point is labeled with a corresponding color family label BRN, ORG, PNK, RD, PRP, as depicted in scatter plot 302. At this step, labels are "manually" assigned to each point in the database by a human color expert based on the subjective perception of makeup colors. These labels may correspond to assessments of brown family BRN, orange family ORG, pink family PNK, red family RD, and purple family PRP. This "manual" assignment is performed once in order to configure the computer-implemented classification method based on a specific subjective perception of makeup colors. This one-time manual assignment may be performed based on the following in conjunction with Figure 8 The conventional makeup color classification technology is used to perform the above-mentioned

[0088] Second, mathematical calculations are performed to extrapolate continuous volumes in L*a*b* space for each discrete color system scatter plot. As shown in Figure 304, the mathematical calculations generate an envelope surrounding all points of each corresponding system, such as through traditional triangulation techniques (such as Delaunay triangulation) and alpha shape generation.

[0089] Additionally, the mathematical calculations advantageously perform interpolation of the extended envelopes, as shown in Figure 306. The extensions are configured to fill in gaps between the colorway volumes until the corresponding opposing surfaces of adjacent envelopes match each other. This can be performed, for example, by selecting the nearest neighbor based on the distance to the triangulated surface of each colorway.

[0090] When all corresponding opposing surfaces of adjacent envelopes match each other with no empty gaps between them, the envelope obtained at graph 306 defines the enclosure of the corresponding color system volume.

[0091] Thus, an input point located in the L*a*b* color space according to its coordinates is assigned a corresponding label BRN, ORG, PNK, RD, PRP of the envelope surrounding the input point in the L*a*b* color space.

[0092] Figure 4 Identification of the location of the input coordinate data 104 relative to at least one luma boundary value BndVal1 , BndVal2 is shown, which is used to assign a luma subsystem label at step 108 of the method 100 .

[0093] In this example, the brightness sub-system labels include a "bright" brightness label LGT, a "medium" brightness label MDM, and a "dark" brightness label DRK, and are identified by comparison with a first brightness boundary value BndVal1 and a second brightness boundary value BndVal2 that is lower than the first brightness boundary value BndVal1. If the brightness input coordinate L* is greater than the first brightness boundary value BndVal1, the bright label LGT is assigned, if the brightness input coordinate L* is between the first brightness boundary value BndVal1 and the second brightness boundary value BndVal2, the medium label MDM is assigned, and if the brightness input coordinate L* is lower than the second brightness boundary value BndVal2, the dark label DRK is assigned.

[0094] Here too, both brightness boundary values ​​BndVal1, BndVal2 are specially designed based on the subjective perception of makeup color, and in particular, the level of the brightness boundary value on the L* axis can be set according to the chromaticity of the input coordinate data.

[0095] In addition, for input coordinates whose chromaticity value C* is greater than a threshold selected according to the subjective perception of the makeup color, the brightness boundary values ​​BndVal1, BndVal2 are advantageously reduced by a step size C*Stp. In fact, for the colors with the highest chromaticity, in particular for the combination Figure 5 The brightness boundary values ​​BndVal1, BndVal2 are slightly reduced for the "high" chroma subsystem label assigned at step 108 of the method 100. This advantageously allows compensation for an effect known as the Helmholtz Kohlrausch effect, in which the subjective perception of brightness increases with increasing chroma.

[0096] Figure 5 Identification of the position of the input coordinate data 104 relative to at least one chroma boundary line BndCrcl1 , BndCrcl2 in the a*b* plane including the input coordinate point is shown, which is used to assign a chroma subsystem label at step 108 of the method 100 .

[0097] In this example, the chroma subsystem labels include a "high" chroma label HGH, a "middle" chroma label INTR, and a "low" chroma label LW, and are identified by comparison with a first chroma boundary line BndCrcl1 and a second chroma boundary line BndCrcl2 that is lower than the first chroma boundary line BndCrcl1. In the a*b* plane, the chroma level C* is defined by the Euclidean distance of a point from the origin (0; 0), that is, Or C*=(a*2+b*2) 1 / 2 Therefore, the first chromaticity boundary line BndCrcl1 and the second chromaticity boundary line BndCrcl2 are circular, ie, the lines have a circular appearance.

[0098] If the chromaticity of the input coordinate C* is greater than the first chromaticity boundary line BndCrcl1, a high label HGH is assigned, if the chromaticity of the input coordinate C* is between the first chromaticity boundary line BndCrcl1 and the second chromaticity boundary line BndCrcl2, an intermediate label INTR is assigned, and if the chromaticity of the input coordinate C* is less than the second chromaticity boundary line BndCrcl2, a low label LW is assigned.

[0099] Here too, both chromaticity boundary lines BndCrcl1 and BndCrcl2 are specially designed based on the subjective perception of makeup colors, and in particular, these lines are defined only for the hues in the a*b* plane that are susceptible to the application of makeup products. In the example of lipstick makeup products, the hues susceptible to application are approximately located on the half-plane of positive values ​​on a*, i.e., from the yellow-orange hue yORG to the blue-purple hue bPRP.

[0100] In addition, the chromaticity border lines BndCrcl1, BndCrcl2 advantageously depend on the hue of the input coordinate data in order to take into account the subjective perception of the chromaticity according to the hue of the corresponding color. In fact, for example, the orange hue yORG appears "weaker" in terms of chromaticity than the purple hue bPRP. Therefore, the two chromaticity border lines BndCrcl1, BndCrcl2 may have the appearance of a spiral portion with a slightly larger radius on the positive values ​​of the b* side (yellow-orange hue yORG) and a slightly narrower radius on the negative values ​​of the b* side (blue-violet hue bPRP) compared to the spiral radius around the zero value of b* (red hue at positive values ​​of a*).

[0101] Figure 6 The input coordinate data 104 is shown relative to the color system capacity ( Figure 3 ) is used to assign a hue subset label at step 108 of method 100.

[0102] In this example, the hue sub-system labels include a "warm" hue label WRM, a "neutral" hue label NTR, and a "cold" hue label CL, and are identified by comparison with the first hue boundary surface BndSrfc1 and the second hue boundary surface BndSrfc2, thereby defining the space within each corresponding hue capacity BRN, ORG, RD, PNK, PRP.

[0103] If the input point is located on one side of the first hue boundary surface BndSrfc1, a warm label WRM is assigned, if the input point is located between the other side of the first hue boundary surface BndSrfc1 and one side of the second hue boundary surface BndSrfc2, a neutral label NTR is assigned, and if the input point is located on the other side of the second hue boundary surface BndSrfc2, a cold label CL is assigned.

[0104] Here too, both hue boundary surfaces BndSrfc1, BndSrfc2 are specially designed based on subjective perception of makeup colors, and in particular, the positions and neutral ranges of these surfaces are determined by the visual results of makeup color experts. The hue boundary surfaces BndSrfc1, BndSrfc2 are defined along both lightness (e.g., from L*=20 to 90) and chromaticity (e.g., from C*=0 to the maximum level), and their coordinate definitions can be registered in a lookup table for performing identification of the relative positions of input coordinate data.

[0105] Figure 7 Shows the above combination Figures 1 to 6 Example results at the final step 110 of the classification method 100 described.

[0106] At a final step 110, the cosmetic product colors are advantageously categorized according to the assigned labels in an identification card diagram shown, which depicts a hierarchical arrangement of colors first according to color family labels, secondly according to hue subfamily labels, and then also according to lightness subfamily labels and chroma subfamily labels.

[0107] In the identification card shown, a variety of cosmetic product colors are classified together, and each color is combined with the above Figures 1 to 6 The method is processed to assign corresponding labels that define the position of the color in the identification card image.

[0108] In the identification card diagram, the colors are arranged accordingly in a table organized by a main row representing each color family label "brown", "pink", "orange", "purple" and "red" and a main column representing each hue subfamily label "cool", "neutral" and "warm". In addition, each main row includes a set of sub-rows for the corresponding lightness subfamily labels "light", "medium" and "dark", and each main column includes a set of sub-columns for the corresponding chromaticity subfamily labels "low", "middle" and "high".

[0109] Alternatively, the table may be organized without subcolumns for these groups of labels for each chromaticity subline, and instead the colors in each luma subrow may be ordered in ascending order of chromaticity value, e.g., ascending order from left to right in the corresponding subrow.

[0110] Since the colors are located according to their labels in the identification card diagram (ie, according to the subjective perception they evoke as cosmetic products), the color expert can perform an improved analysis of the classified set of cosmetic colors.

[0111] In fact, firstly, since the classification is performed automatically by a computer, identification cards can be generated multiple times for different choices of makeup product color groups in a very fast manner.

[0112] Therefore, in contrast to traditional techniques for this classification (manually performed by visually inspecting thumbnail color samples, e.g. Figure 8 This approach can save color experts a lot of time compared to the previous method.

[0113] Second, since the positions of the colors correspond to the subjective perception they evoke as a cosmetic product, the identification card is not limited by the color display performance of a given computer monitor. In fact, a computer monitor may display the same color for two different data, for example, a low-chroma warm pink may appear the same as a low-chroma cool brown on a computer monitor, because of its limited performance in terms of "color number" compared to the sensitivity of human eye perception. This situation, without the classification in the identification card, would lead to an erroneous analysis of these colors. However, the positions of the two colors in the identification card provide the color expert with information about their actual difference, despite the fact that the two colors appear with the same signal on the computer monitor.

[0114] Figure 8 Photograph of a traditional technique for classifying the colors of cosmetic products, for example, in order to compare products of different brands and franchises of the same holding company or between own traditional products and those of competitors.

[0115] This classification is usually done by a color expert using physical samples of the color as perceived by the human eye, such as thumbnail color samples of a makeup product that are individually characterized based on subjective perception and spread out on a lab bench or whiteboard.

[0116] This traditional technique is obviously time consuming and is not suitable or even possible to perform multiple times for many different sets of makeup colors; whereas, as described above, the identification card image can be easily generated any number of times on the fly as needed.

[0117] Fig. 9 An example of a classical computer representation of a set of colors is shown, here a diagram projected on a 2-dimensional a*b* plane (i.e. ignoring the L* coordinate). Even though the CIELAB a*b* plane was designed to provide a close representation of human color perception, this representation cannot be used to analyze makeup colors. In fact, even though the identification of the color families "brown", "orange", "pink", "red", "purple" can be partially performed visually by a color expert on such a diagram, other representations are limited to the color display performance of a computer screen.

[0118] The aforementioned problem still exists that the same color signal is displayed for two different color data on a computer monitor (e.g., a low chroma warm pink may also appear as a low chroma cool brown), and classification based on subjective perception of makeup color is not feasible.

[0119] From another perspective, although the chromaticity value can be viewed as the distance of a color point from the origin, the subjective perception of chromaticity varies according to the hue. Therefore, a high chromaticity point in a given hue (e.g., a purple hue) may paradoxically be plotted closer to the origin than a lower chromaticity point in another hue (e.g., an orange hue), causing the color points to appear mixed and disordered in the a*b* diagram. Classification based on subjective perception of makeup color still needs to be performed visually, so it is not feasible due to the limitations of the display performance of computer monitors.

[0120] Fig.10 A complete computer tool is presented for assisting color experts (i.e., "users") in developing colors for cosmetic products, which tool benefits from the combination of the above Figures 1 to 7 The classification method and identification card diagram.

[0121] The computer-implemented tool controlled by the user may be embodied in practice as a computer program product comprising instructions, which, when executed by a computer, cause the computer to perform the tool according to the present disclosure; or may be embodied in practice as a computer-readable storage medium comprising instructions, which, when executed by a computer, cause the computer to perform the tool according to the present disclosure.

[0122] The tool first includes a drawing mode configured to provide a selection of makeup colors as described above. Figure 7 The identification card diagram.

[0123] Thus, a user may be able to select a set of cosmetic product colors from a cosmetic product color library (eg, a library including in-house brand and franchise cosmetic colors and / or competitor brand and franchise cosmetic colors) for display in a graphical mode.

[0124] Users can also select the target market for the makeup product or choose from many other options based on business sectors from the color library.

[0125] exist Figure 7 In the example of , the identification card chart thus categorizes the selected set of colors in a table that is organized as follows: the five main rows represent the brown, pink, orange, purple, and red color family labels; the three main columns represent the cool, neutral, and warm sub-family labels; the three sub-rows of each main row represent the light, medium, and dark brightness sub-family labels; and the three sub-columns of each main column represent the low, middle, and high chroma sub-family labels.

[0126] Therefore, the tool’s graphing mode provides the user with a data visualization of instrument measurement data arranged in a table based on subjective perception of makeup colors. Graphing allows the user to select shades to visualize in this meaningful identity card diagram, for example comparing a competitor’s successful shade range with the company’s internal shade range to understand shade gaps that are not currently covered by the company’s internal brands.

[0127] Fig.11 , Fig.12 and Fig.13 The application mode of the tool is demonstrated, suitable for digitally simulating an on-face visualization of the applied tonal range to more specifically consider the makeup color selected by the user.

[0128] The application mode is suitable for the user to select a set of one or more makeup product colors from the identification card image displayed in the drawing mode and select a skin color photo model or several skin color photo models. The application mode is configured to display a simulated image of applying the selected set of makeup product colors on at least one skin color photo model, respectively.

[0129] Advantageously, the simulation of applying a selected set of makeup product colors on at least one skin color photo model, respectively, is performed by a pseudo-spectral imaging system disclosed in the scientific publication "Development of a Multispectral Imaging System Using an RGB Camera Under Two Illuminations" by Liu Z., Xiao K., Pointer M. et al. (Proceedings of the 28th IS&T Color and Imaging Conference, 28th Color & Imaging Conference, November 4-19, 2020, online, Society for Imaging Science and Technology); or by the technology disclosed in U.S. Granted Patent Document No. US 8,498,456 B2 (entitled "Methods and Systems for Applying Cosmetic and / or Accessory Enhancements to Digital Images"); or by the technology disclosed in U.S. Patent Application Document No. US2018 / 0075524 A1 ("Applying Virtual Makeup Products").

[0130] Fig.11 Possible options in the application mode are shown. In this example, the user has selected about nine or more shades from the drawing mode, and the user has selected a skin tone photo model.

[0131] Simulated images of the selected set of makeup product shades applied on the selected skin tone photo model are displayed in a grid that dynamically adjusts based on the selection. In this example, the grid is 3 x 3 images and can be scrolled up and down to navigate the grid.

[0132] Two view options are available to display the simulated image, for example, full face view (see Fig.12 ) or Fig.11 A close-up view depicted.

[0133] The tone selection can be modified at any time by the same process as in drawing mode, such as using a check box for each simulated image. In addition, the modified tone selection can be automatically reflected in drawing mode.

[0134] Fig.12 Other possible options in the application mode are shown. In this example, the user switches from the drawing mode or a previous application mode (e.g., from Fig.11 In the example of the application mode of FIG, two tones are selected and the user selects two skin color photo models.

[0135] Fig.13 Other possible options in the application mode are shown. In this example, the user switches from the drawing mode or a previous application mode (e.g., from Fig.11 ) in the example of the application mode of FIG. 2 , two tones are selected, and the user selects a skin tone photo model. When only two tones are selected, the application mode provides a side-by-side comparison mode in which dragging the cursor left and right allows the two tones to be compared immediately.

[0136] Fig.14 The tool’s color creation mode is demonstrated, which is suitable for digitally creating new shades and visualizing them on a simulated face of the shade during the creation process, as a preliminary design step in creating a new makeup product.

[0137] The color creation mode is suitable for the user to select at least one skin color photo model and set parameters for generating a custom color. The color creation mode is configured to display a simulated image of applying the custom color on the at least one skin color photo model in real time.

[0138] For example, the parameters for generating a custom color may include a hue parameter, a saturation parameter, and a brightness parameter. A gloss parameter such as "matte", "satin", or "glossy" may also be provided.

[0139] In addition, the color creation mode is advantageously adapted for the user to select an existing cosmetic product color from the displayed identification card image, and is configured to simultaneously display simulated comparison images of the application of the custom color and the selected set of cosmetic product colors, respectively, on at least one skin color photo model. For example, the simultaneous display of the comparison images can be used as shown in FIG. Fig.13 The drag cursor depicted or Fig.11 or Fig.12 The simulated image grid depicted is implemented.

[0140] For example, a user may compare two shades digitally applied to a face by viewing the two shades in a side-by-side view and additionally using the color difference indicators Δ(E), L*, C*, h(θ).

[0141] In summary, combined with the above Figures 10 to 14 The described tool provides a digital system that enables instant analysis of color cosmetic shades and creation of shade ranges based on a comprehensive process of user interaction according to:

[0142] 1) Data visualization representing specific subjective perceptions of makeup colors, which allows users to select shades to visualize and conduct meaningful analysis of products and markets.

[0143] 2) Digital application of measured hues on images depicting models of different skin tones captured using a pseudo-spectral imaging system, which allows users to visually evaluate hues on models of different skin types for market suitability by digitally applying hues (e.g., lipstick hues) on images showing faces of models with different skin tones.

[0144] 3) The ability to accurately conceive and apply color tones digitally on images captured using the aforementioned pseudo-spectral analog imaging system, which allows the user to digitally create color tones using inputs of brightness, chroma, hue, and makeup gloss values, which will be rendered in real time on the image of the model's face.

Claims

1. A computer-implemented method (100) for automatically classifying the colors of a cosmetic product (102), the method comprising: - providing input coordinate data of the color of the cosmetic product in the L*a*b* space (104); - assigning a color family label (106) to the cosmetic product color based on identifying a color family volume in a set of color family volumes (ORG, PNK, RD, PRP, BRN) in the L*a*b* space that contains the input coordinate data; The color system capacity (ORG, PNK, RD, PRP, BRN) is designed based on the subjective perception of makeup colors, and the makeup product colors are classified according to the assigned labels.

2. The method according to claim 1, wherein: The set of color family volumes (ORG, PNK, RD, PRP, BRN) in the L*a*b* space is generated from a database (302) comprising a finite number of points in the L*a*b* space, each point being labeled with a corresponding family label according to a subjective perception of a makeup color, and by mathematical calculations including triangulation (304) of envelopes surrounding all points of each corresponding family and interpolation (306) of expanding the envelopes until corresponding relative surfaces of adjacent envelopes match each other, the envelopes defining an enclosing circle of the corresponding color family volume.

3. The method according to any one of claims 1 or 2, wherein: The color system volumes (BRN, PNK, ORG, PRP, RD) are configured to define brown, pink, orange, purple and red in the L*a*b* space based on the subjective perception of makeup color.

4. The method according to any one of the preceding claims, further comprising: - assigning a brightness sub-label (108) to the cosmetic product color based on the identification of the position of the input coordinate data relative to at least one brightness boundary value (BndVal1, BndVal2) in the brightness coordinate axis L* of the L*a*b* space; The at least one brightness boundary value (BndVal1, BndVal2) is designed based on a subjective perception of a makeup color, and the makeup product color is classified based on the assigned label.

5. The method according to claim 4, wherein: For input coordinates whose chromaticity values ​​are greater than a threshold set according to subjective perception of makeup color, the at least one brightness boundary value (BndVal1, BndVal2) in the brightness coordinate axis L* of the L*a*b* space is reduced by a step size (C*Stp).

6. The method according to any one of claims 4 or 5, wherein: The luma subsystem includes a bright label (LGT) assigned when the luma input coordinate is greater than a first luma boundary value (BndVal1), a medium label (MDM) assigned when the luma input coordinate is between the first luma boundary value (BndVal1) and a second luma boundary value (BndVal2), and a dark label (DRK) assigned when the luma input coordinate is lower than the second luma boundary value (BndVal2).

7. The method according to any one of the preceding claims, further comprising: - assigning a chromaticity subsystem label (108) to the cosmetic product color based on the position of the input coordinate data relative to at least one chromaticity boundary line (BndCrcl1, BndCrcl2) in the a*b* plane of the L*a*b* space; The at least one chromaticity boundary line (BndCrcl1, BndCrcl2) is designed based on subjective perception of makeup color, and the makeup product color is classified according to the assigned label.

8. The method according to claim 7, wherein: The at least one chromaticity boundary line (BndCrcl1, BndCrcl2) in the a*b* plane of the L*a*b* space varies according to the hue of the input coordinate data based on subjective perception of makeup color.

9. The method according to any one of claims 7 or 8, wherein: The chromaticity subsystem includes a high label (HGH) assigned when the chromaticity of the input coordinates is greater than a first chromaticity boundary line (BndCrcl1), an intermediate label (INTR) assigned when the chromaticity of the input coordinates is between the first chromaticity boundary line (BndCrcl1) and the second chromaticity boundary line (BndCrcl2), and a low label (LW) assigned when the chromaticity of the input coordinates is lower than the second chromaticity boundary line (BndCrcl2).

10. The method according to any one of the preceding claims, further comprising: - assigning a hue sub-family label (108) to the cosmetic product color based on the position of the input coordinate data relative to at least one hue boundary surface (BndSrfc1, BndSrfc2) within the corresponding hue volume in the L*a*b* space; The at least one hue boundary surface (BndSrfc1, BndSrfc2) is designed based on subjective perception of makeup colors, and the makeup product colors are classified based on the assigned labels.

11. The method according to claim 10, wherein: The at least one hue boundary surface (BndSrfc1, BndSrfc2) in each color system volume of the L*a*b* space is defined based on subjective perception of makeup color.

12. The method according to any one of claims 10 or 11, wherein: The hue subset includes a warm label (WRM) assigned when the input coordinates are located on one side of a first hue boundary surface (BndSrfc1), a neutral label (NTR) assigned when the input coordinates are located between the other side of the first hue boundary surface (BndSrfc1) and one side of a second hue boundary surface (BndSrfc2), and a cold label (CL) assigned when the input coordinates are located on the other side of the second hue boundary surface (BndSrfc2).

13. A method according to any one of the preceding claims in combination with claims 4, 7 and 10, wherein: The cosmetic product colors are hierarchically classified first according to the color family label, second according to the hue sub-family label, and then according to the lightness sub-family label and the chroma sub-family label.

14. A computer-implemented tool, intended to be controlled by a user, for aiding in cosmetic product color development, the tool comprising: - A drawing mode (Figure 10), which is suitable for a user to select a set of cosmetic product colors from a cosmetic product color library, is configured to classify each of the colors using the method for classifying cosmetic product colors according to any one of claims 1 to 13, and is configured to display a graph of the selected colors arranged in a table according to respectively assigned labels, the table being organized by a main row representing each color system label and a main column that may represent a selected set of each hue sub-system label, each main row may include a sub-row representing each brightness sub-system label, and each main column may include a sub-column representing each chromaticity sub-system label.

15. The tool according to claim 14, wherein: The method for classifying the colors of cosmetic products is according to any one of claims 1 to 13 in combination with claims 4, 7 and 10, the table being organized in the following manner: - The five main rows represent the brown, pink, orange, purple and red labels respectively. - The three main columns represent the cool, neutral, and warm sub-family labels, - the three sub-rows of each main row represent the light, medium, and dark brightness sub-line labels, respectively, and - The three sub-columns of each main column represent the low, middle and high chroma sub-line labels respectively.

16. A tool according to any one of claims 14 or 15, further comprising: - Application mode (Figures 11, 12, and 13), which is suitable for the user to select a set of at least one makeup product color from the displayed picture (ID card) and select at least one skin color photo model, and is configured to display simulated images of applying the selected set of makeup product colors on the at least one skin color photo model.

17. A tool according to any one of claims 14 to 16, further comprising: - A color creation mode (FIG. 14), adapted for a user to select at least one skin color photo model and set parameters for generating a custom color, configured to display a simulated image with the custom color applied on the at least one skin color photo model.

18. The tool according to claim 17, wherein: The color creation mode (Figure 14) is further adapted for a user to select a set of at least one cosmetic product color from the displayed diagram, and is configured to simultaneously display simulated comparison images of the custom color and the selected set of cosmetic product colors applied respectively on the at least one skin color photo model.

19. A computer program product comprising instructions which, when the program is executed by a computer, cause the computer to perform the method according to any one of claims 1 to 13, or cause the computer to perform the tool according to any one of claims 14 to 18.

20. A computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to perform the method according to any one of claims 1 to 13, or cause the computer to perform the tool according to any one of claims 14 to 18.

Images