Method for classifying the color of a makeup product and tool to aid in the development of makeup product colors

Abstract

Makeup product colour classification method and makeup product colour development aid tool The computer-implemented tool, intended to be controlled by a user, to aid in the development of makeup product colours, includes: - a mapping mode (Figure 10) adapted for the user to select a group of makeup product colours from a makeup product colour bank, configured to classify each colour in the selected colour group with the makeup product colour classification process according to labels designed according to a subjective perception of makeup colours.The mapping mode is configured to display a map of the selected colors arranged in a table according to their assigned labels. The table is organized into a main row for each color family label, and a main column for each hue subfamily label. Each main row includes a sub-row for each brightness subfamily label, and each main column includes a sub-column for each chroma subfamily label. Figure for the abbreviation: Fig. 10.

Description

Title of the invention: Method for classifying a color of a makeup product and tool for assisting in the development of colors of makeup products

[0001] Embodiments of the invention relate to a method for classifying a color of a cosmetic makeup product, which is implemented automatically by a computer, and a tool for assisting in developing colors of cosmetic makeup products, which is implemented by a computer and controlled by a user.

[0002] By "cosmetic product" is meant any product as defined in Regulation (EC) No 1223 / 2009 of the European Parliament and of the Council of 30 November 2009 concerning cosmetic products. A cosmetic makeup product or "makeup product" is more particularly intended to cover a surface of the body in order to modify the perceived color and / or texture.

[0003] Developing makeup product colors may include viewing and comparing existing makeup product colors in a given market, and for different skin tones, to identify, for example, trends in customer preferred color shades, or to identify color shades that may be lacking and needed in the given market.

[0004] For example, the markets studied can range from cheap products to luxury products, and the products can vary depending on the location of the market, the Asian or Chinese market being for example different from the European market.

[0005] The comparison between makeup products can be made between products of different brands and franchises of the same company, or between existing products owned and products of a competitor.

[0006] Visualization is usually performed by color experts with physical samples of the colors under the perception of the human eye, for example with miniature color samples of makeup products, because of the strong reaction in the subjective perception of makeup colors that can be produced by a slight variation in the absolute color. Indeed, in the example of a red color for a lipstick, a first red color may subjectively appear cold and lifeless while a second red color, very close in absolute terms to the first red color, may subjectively appear warm and rich.

[0007] The absolute difference between two colors can be, for example, the distance separating these colors in a “standard observer” model, such as the “CIEXYZ” color space or the “CIELAB” color space. These models Standard observer colors are defined by the International Commission on Illumination (CIE), and the colors they define are not relative to any particular device such as a computer monitor or printer, but rather relate to the CIE Standard Observer, which is an 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 primary colors red, green, and blue, and is generally used in electronics to detect and display colors.

[0009] The "CIELAB" color space, also called L*a*b*, expresses colors as three coordinates: L* for lightness, and a* and b* for greenish to reddish chromaticity and bluish to yellowish chromaticity respectively (chromaticity conveying both hue and chroma measurements). In the a*b* plane, a single hue can be identified by a single angle in the trigonometric circle. The CIELAB color space is designed to be more perceptually uniform, for example, than the RGB color space. However, the subjective perception of makeup product colors is not well translated into the CIELAB color space, so a slight variation in the absolute color in the CIELAB color space can still produce a strong reaction in the subjective perception of makeup colors.

[0010] The perception of colors by the human eye goes beyond L*a*b* coordinates and attaches importance to subjective perceptions such as the "background effect" of a color, where for example looking at a single colored square surrounded by a colored background gives a different perception of the colored square depending on the color of the background. With regard to makeup products, the perception of a lipstick color can vary depending on the skin tone of the person wearing it.

[0011] Therefore, it is difficult to provide a computerized tool to assist color experts in developing makeup product colors, due to inconsistencies in the digital transcription of a color, such as RGB or L*a*b* coordinates, versus the subjective perception of makeup colors.

[0012] Furthermore, another difficulty encountered with the computerized tool to assist color experts in developing makeup product colors is that computer screens are not capable of displaying the same range of colors in the same way and are in any case all limited in the range of colors they can display (generally called "number of colors"). The resulting problem is that two different colors, which would have distinguishable properties under the perception of the human eye, can be displayed in exactly the same way on different screens, and therefore be unusable for an analysis by a makeup color expert.

[0013] Thus, visualizations and comparisons of colors of makeup products are generally always carried out with miniature color samples of the makeup products, often in very large numbers, in laboratory conditions, and are not achievable remotely with a computer, for example in home working conditions.

[0014] Another difficulty in developing colors of makeup products is that the applied colors of the product may vary depending on the skin tone of the user wearing the makeup, and the subjective perception of the color may also vary depending on the skin tone of the user. For example, the "background effect" of skin tone, as mentioned above, may change the perception of a makeup product color, and, in principle, a light or pale makeup color may become darker when applied to dark skin, and a dark and rich makeup color may appear excessively intense when applied to light skin, while appearing perfectly matched to dark skin.

[0015] Therefore, the development of makeup product colors requires testing phases on human models to analyze the actual subjective perception when the makeup is worn. These testing phases are generally long and also expensive.

[0016] There is therefore a need to provide a tool, particularly a computerized tool, to assist color experts and save their time in developing makeup product colors, which would be adapted in terms of subjective perception of makeup colors, which would not be limited to the performance of a computer screen, such as "number of colors", and which would allow the development of makeup colors suitable for a variety of skin tones.

[0017] According to one aspect of the invention, a computer-implemented method for automatically classifying a color of a makeup product, comprises: - the provision of input coordinate data in the L*a*b* space of a makeup product color; - assigning a color family label (e.g., brown, pink, orange, purple, red) to the makeup product color based on an identification of a color family volume, from a set of color family volumes in L*a*b* space, containing the input coordinate data; the color family volumes being designed according to a subjective perception of makeup colors, and the color of the makeup product being classified according to the assigned labels.

[0018] Said subjective perception of makeup colors used for design volumes, and the boundary value, line and area defined below are advantageously defined by human color experts, for example according to the aforementioned sensitive subjective perception specific to makeup colors. Advantageously, the human color experts can provide a visual database of discrete color points, and the design of the volume and boundary value, line and area can be performed to achieve continuity of the color points by a data-driven calculation configured to fit the visual database. For example, the data-driven calculation configured to fit the visual database can be implemented by a model trained in machine learning.

[0019] In other words, the computer-implemented method according to this aspect makes it possible to classify colors on the basis of labels that are representative of subjective perceptions of makeup colors, and defined by conditions established by the subjective perceptions specific to the makeup colors. The resulting classification, for example applied to each color of a group of analyzed makeup products, can thus provide useful information to a color expert developing makeup product colors, despite the limitations of computer representations of colors (color spaces) and of the display of colors on the screen.

[0020] According to one embodiment, the set of color family volumes in the L*a*b* space is generated from a database including a finite number of points in the L*a*b* space, each point being labeled with a respective family label according to the subjective perception of makeup colors, and from a mathematical calculation comprising a triangulation generating envelopes encompassing all points of each respective family and an interpolation spreading the envelopes until the respective facing surfaces of neighboring envelopes correspond to each other, the envelopes defining the enclosure of the respective color family volumes.

[0021] According to one embodiment, the color family volumes are configured to delineate the colors brown, pink, orange, violet and red in the L*a*b* space according to the subjective perception of makeup colors.

[0022] According to one embodiment, the method further comprises: - assigning a brightness subfamily label (e.g., light, medium, dark) to the makeup product color based on an identification of the position of the input coordinate data relative to at least one brightness boundary value in the brightness coordinate axis L* of the L*a*b* space; the at least one brightness limit value being designed according to a subjective perception of makeup colors, and the makeup product color being classified according to the assigned labels.

[0023] According to one embodiment, the at least one brightness limit value in the brightness coordinate axis L* of the L*a*b* space is reduced by one step for the input coordinates having a chroma value greater than a threshold defined according to the subjective perception of makeup colors.

[0024] According to one embodiment, the brightness subfamily includes a light label assigned if the brightness input coordinate is greater than a first brightness limit value, a medium label assigned if the brightness input coordinate is between the first brightness limit value and a second brightness limit value, and a dark label assigned if the brightness input coordinate is less than the second brightness limit value.

[0025] According to one embodiment, the method further comprises: - assigning a chroma subfamily label (e.g., high, intermediate, low) to the makeup product color based on the position of the input coordinate data relative to at least one chroma boundary line in an a*b* plane of L*a*b* space; the at least one chroma limit line being designed according to a subjective perception of makeup colors, and the makeup product color being classified according to the assigned labels.

[0026] According to one embodiment, the at least one chroma boundary line in the a*b* plane of the L*a*b* space varies as a function of the hue of the input coordinate data, according to the subjective perception of makeup colors.

[0027] According to one embodiment, the chroma subfamily includes a high label assigned if the chroma of the input coordinates is greater than a first chroma boundary line, an intermediate label assigned if the chroma of the input coordinates is between the first chroma boundary line and a second chroma boundary line, and a low label assigned if the chroma of the input coordinates is less than the second chroma boundary line.

[0028] According to one embodiment, the method further comprises: - assigning a hue tone subfamily label (e.g., cool, neutral, warm) to the makeup product color according to the position of the input coordinate data relative to at least one hue tone boundary surface within the respective color family volume in L*a*b* space; the at least one shade tone boundary surface being designed according to a subjective perception of makeup colors, and the makeup product color being classified according to the assigned labels.

[0029] According to one embodiment, the at least one hue tone boundary surface in each color family volume of the L*a*b* space is defined according to the subjective perception of makeup colors.

[0030] According to one embodiment, the hue tone subfamily includes a warm label assigned if the input coordinates lie on one side of a first hue tone boundary surface, a neutral label assigned if the input coordinates lie between the other side of the first hue tone boundary surface and one side of a second hue tone boundary surface, and a cool label assigned if the input coordinates lie on the other side of the second hue tone boundary surface.

[0031] According to one embodiment, the makeup product color is hierarchically classified first according to the color family label, then according to the hue tone subfamily label, and then according to the brightness subfamily label and the chroma subfamily label.

[0032] This hierarchy makes it possible to present a practical classification of the labels, for example for the analysis of colors of makeup products. That being said, another hierarchy can also be used and, with regard to the processing of assigning the labels, the respective processing step can be carried out at the same time or possibly the brightness label must be processed after the chroma label, since the brightness limit line values ​​can depend on the chroma label.

[0033] According to another aspect, a computer-implemented tool, intended to be controlled by a user, for assisting in the development of makeup product colors, comprises: - a mapping mode adapted for the user to select a makeup product color group from a makeup product color bank, configured to classify each color of the selected color group with the method of classifying a makeup product color as defined above, and configured to display a map of the selected colors arranged in a table according to the respectively assigned labels, the table being organized into a main row for each color family label, and a main column for each hue tone subfamily label, each main row including a sub-row for each brightness subfamily label, each main column including a sub-column for each chroma subfamily label.

[0034] According to one embodiment, the table is organized into: - five main lines for family labels in brown, pink, orange, purple and red colors respectively, - three main columns for cool, neutral and warm hue tone subfamily labels respectively, - three sub-lines for each main line for sub-family labels of light, medium and dark brightness respectively, and - three sub-columns for each main column for low, medium and high chroma sub-family labels respectively.

[0035] According to one embodiment, the tool further comprises: - an application mode adapted for the user to select a set of at least one makeup product color from the displayed card (ID card) and select at least one complexion photography model, configured to display simulation images of applications of the selected set of makeup product colors respectively on the at least one complexion photography model.

[0036] According to one embodiment, the tool further comprises: - a color creation mode adapted for the user to select at least one skin tone photograph model and define parameters to generate a custom color, configured to display a simulation image of an application of the custom color on the at least one skin tone photograph model.

[0037] According to one embodiment, the color creation mode is further adapted for the user to select a set of at least one makeup product color from the displayed card (ID card), and is configured to simultaneously display comparative images of simulation of applications of the personalized color and the selected set of makeup product colors, respectively on the at least one complexion photograph model.

[0038] According to another aspect, a computer program product comprises instructions which, when the program is executed by a computer, cause the computer to implement the method as defined above, or cause the computer to implement the tool as defined above.

[0039] According to another aspect, a computer-readable storage medium comprises instructions which, when executed by a computer, cause the computer to implement the method as defined above, or cause the computer to implement the tool as defined above.

[0040] In other words, aspects and embodiments particularly provide a digital system that enables instant analysis of cosmetic color shades and shade range creation, based on an integrated process of 1) visualizing instrumental measurement data according to labels suitable for makeup product color classification; 2) digitally applying measured shades to images depicting models of different skin tones, taken for example using a pseudo-spectral imaging system; 3) functionality to accurately digitally design and apply shades to images taken, for example, using the above pseudo-spectral camera system. mentioned.

[0041] It is therefore a defined method and system (tool) which makes it possible to integrate the development of a range of shades, which takes into account in particular the evaluation of makeup on different skin tones, intended for makeup product development teams using a digital platform.

[0042] This allows product development teams to easily create shade ranges, by providing the built-in ability to visualize shades from a database of measured shades and digitally design new shades, to create better performing, skin tone-matching shade ranges for color cosmetic products.

[0043] Furthermore, this increases the efficiency of shade range development, potentially reducing shade range development time by approximately 25% through the ability to digitally visualize and compare lipstick shades in dedicated classification and on-face simulations.

[0044] Additionally, since the aspects and embodiments defined above allow for the development of shade ranges deemed suitable for a wide variety of skin tones, this could reduce the annual number of shades created, since development teams can view and select shades corresponding to specific skin tones from the database.

[0045] Other advantages and specifications of the invention will appear when examining the detailed description of embodiments, in no way limiting, and in relation to the appended drawings, among which:

[0046] [Fig. 1];

[0047] [Fig.2];

[0048] [Fig.3];

[0049] [Fig.4];

[0050] [Fig.5];

[0051] [Fig.6];

[0052] [Fig.7] illustrates embodiments of a method for classifying a color of a makeup product, according to the invention;

[0053] [Fig.8];

[0054] [Fig.9] illustrates difficulties encountered in conventional techniques;

[0055] [Fig. 10];

[0056] [Fig. 11];

[0057] [Fig. 12];

[0058] [Fig. 13];

[0059] [Fig. 14] illustrates embodiments of a development assistance tool for colors of makeup products, according to the invention.

[0060] [Fig.l] illustrates a method 100 for classifying a color of a makeup product 102, which is designed to be implemented automatically by a computer. The makeup product 102 is preferably a lipstick. The method is described herein as providing classification labels for a single color 104, of a single makeup product 102. For example, the makeup product 102 is selected from a database of makeup product colors at a step 102. However, it will be understood that the classification is primarily intended to be applied to multiple colors of multiple makeup products, in order to distinguish these colors according to the subjective human perception of makeup colors. In this case, the method is applied to each color of the set of multiple makeup products.

[0061] The method comprises, at a step 104, providing input coordinate data in the L*a*b* space of the makeup product color 102. As mentioned above and as described below in relation to [Fig. 2], the L*a*b* space is the conventional “CIELAB” color space, expressing colors in the form of three coordinates: L* for brightness, and a* and b* for greenish to reddish chromaticity and bluish to yellowish chromaticity respectively (chromaticity conveying both hue and chroma measurements). Note that chroma is also referred to as chroma in the description and in particular in the figures. Step 104 may include a conventional conversion of the color coordinates from any other color space into coordinates in the L*a*b* color space.

[0062] A first assignment step 106 assigns a color family label (which may be, for example, a brown, pink, orange, purple, or red color family label) to the makeup product color 102-104. As discussed below in relation to [Fig. 3], the color family label may be assigned according to an identification of a color family volume, among a set of color family volumes in the L*a*b* space, containing the input coordinate data. In particular, the color family volumes are specially designed according to a subjective perception of makeup colors. For example, and as described below in relation to [Fig. 3], the color family volumes are determined by adaptation to a visual database, advantageously according to a data-driven calculation.

[0063] The second assignment step 108 assigns a brightness subfamily label (which may be, for example, a light, medium, or dark brightness subfamily), a chroma subfamily label (which may be, for example, a high, intermediate, or low chroma subfamily), and a hue tone subfamily label (which may be, for example, a hue tone subfamily cool, neutral or warm) to makeup product color 102-104.

[0064] As described below in connection with [Fig. 4], the brightness subfamily label may be assigned according to an identification of the position of the input coordinate data 104 relative to at least one brightness limit value in the brightness coordinate axis L* of the L*a*b* space. In particular, at least one brightness limit value is specially designed according to a subjective perception of makeup colors.

[0065] As described below in connection with [Fig.5], the chroma subfamily label may be assigned according to the position of the input coordinate data 104 relative to at least one chroma boundary line in an a*b* plane of the L*a*b* space. In particular, the at least one chroma boundary line is specially designed according to a subjective perception of makeup colors.

[0066] As described below in connection with [Fig. 6], the hue tone subfamily label may be assigned according to the position of the input coordinate data 104 relative to at least one hue tone boundary surface within the respective color family volume 106 in the L*a*b* space. In particular, the at least one hue tone boundary surface is specially designed according to a subjective perception of makeup colors.

[0067] In the final step 110, the color of the makeup product is classified according to the assigned labels, for example and advantageously in an identification card, arranged hierarchically as described below in relation to [Fig.7].

[0068] [Fig.2] illustrates an example of a discrete representation of the L*a*b* color space, also called the CIELAB color space. The CIELAB color space expresses colors in the form of three coordinates: L* for brightness, and a* and b* for greenish to reddish hues and bluish to yellowish hues respectively.

[0069] The L*a*b* space allows the hue, chroma, and value of color points to be easily deduced in the L*a*b* coordinate system. The hue of a color point is identified by the trigonometric angle 0 in the a*b* plane including the point, and can thus be expressed as 0=tan '(b* / a*). The chroma 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 thus be expressed as C*=(a*2+b*2)1 / 2 The value of a color point is defined by its brightness coordinate L*.

[0070] [Fig.3] illustrates the generation of the set of color family volumes in L*a*b* space, which are used for the identification of the color family label of the input coordinate data at the first assignment step 106 of the method 100.

[0071] First, a database comprising a finite number of points in the L*a*b* space for each color family is provided, in which each point is labeled with a respective BRN, ORG, PNK, RD, PRP color family label, as illustrated by the scatter plot 302. At this stage, the labels are assigned “manually” to each point in the database, by human color experts and according to the subjective perception of makeup colors. These labels may correspond to the evaluation of a BRN brown color family, an ORG orange color family, a PNK pink color family, a RD red color family and a PRP purple color family. This “manual” assignment is performed only once to configure the computer-implemented classification method according to the specific subjective perception of makeup colors.This unique manual assignment can be done according to a classic makeup color classification technique described below in relation to [Fig.8].

[0072] Next, a mathematical calculation is performed to extrapolate a continuous volume into L*a*b* space for each discrete color family scatterplot. As shown in diagram 304, the mathematical calculation generates envelopes encompassing all points of each respective family, for example, through a conventional triangulation technique, such as Delaunay triangulation, and alpha shape generation.

[0073] Furthermore, the mathematical calculation advantageously performs an interpolation spreading the envelopes, as shown in diagram 306. The spreading is configured to fill the gaps between the color family volumes, until the respective facing surfaces of neighboring envelopes correspond to each other. This can for example be done by choosing the nearest neighbor based on the distance to the triangulation surface of each color family.

[0074] The envelopes obtained in diagram 306, when all the respective facing surfaces of neighboring envelopes correspond to each other without empty space between them, define the enclosure of the respective color family volumes.

[0075] An entry point, positioned in the L*a*b* color space according to its coordinates, is thus assigned to the respective BRN, ORG, PNK, RD, PRP label of the envelope which encompasses the entry point in the L*a*b* color space.

[0076] [Fig.4] illustrates the identification of the position of the input coordinate data 104 relative to at least one brightness boundary value BndVall, BndVal2 used to assign the brightness subfamily label in step 108 of the method 100.

[0077] In this example, the brightness subfamily labels include a “light” brightness label LGT, a “medium” brightness label MDM, and a “dark” brightness label DRK and are identified relative to a first brightness limit value BndVall and a second brightness limit value BndVal2, which is lower than the first brightness limit value BndVall. The light label LGT is assigned if the brightness input coordinate L* is greater than the first brightness limit value BndVall, the medium label MDM is assigned if the brightness input coordinate L* is between the first brightness limit value BndVall and the second brightness limit value BndVal2, and the dark label DRK is assigned if the brightness input coordinate L* is lower than the second brightness limit value BndVal2.

[0078] Again, the two brightness limit values ​​BndVall, BndVal2 are specially designed according to the subjective perception of makeup colors, and in particular, the level in the L* axis of the brightness limit values ​​can be defined according to the chroma of the input coordinate data.

[0079] Furthermore, the brightness limit values ​​BndVall, BndVal2 are advantageously decreased by a step C*Stp for the input coordinates having a chroma value C* greater than a threshold chosen according to the subjective perception of makeup colors. Indeed, the brightness limit values ​​BndVall, BndVal2 decrease slightly for the highest chromatic colors, in particular for the “high” chroma subfamily label assigned in step 108 of the method 100 described in relation to [Fig. 5]. This advantageously makes it possible to compensate for an effect, called the Helmholtz Kohlrausch effect, where the subjective perception of brightness increases with the chroma.

[0080] [Fig.5] illustrates the identification of the position of the input coordinate data 104 relative to at least one chroma boundary line BndCrcll, BndCrcl2 in the a*b* plane including the input coordinate point, used to assign the chroma subfamily label in step 108 of the method 100.

[0081] In this example, the chroma subfamily labels include a "high" chroma label HGH, an "intermediate" chroma label INTR, and a "low" chroma label LW and are identified with respect to a first chroma boundary line BndCrcll and a second chroma boundary line BndCrcl2, lower than the first chroma boundary line BndCrcll. In any plane a*b*, the chroma level C* is defined by the Euclidean distance of the point from the origin (0;0), i.e., Q — ^2 + ^2 , or C*=(a*2+b*2)1 / 2- Therefore, the first chroma boundary line BndCrcll and the second chroma boundary line BndCrcl2 resemble a circle, i.e., these lines have a circular appearance.

[0082] The high label HGH is assigned if the chroma of the input coordinate C* is greater than the first chroma boundary line BndCrcll, the intermediate label INTR is assigned if the chroma of the input coordinate C* is between the first chroma boundary line BndCrcll and the second chroma boundary line BndCrcl2, and the weak label LW is assigned if the chroma of the input coordinate C* is less than the second chroma boundary line BndCrcl2.

[0083] Again, the two chroma boundary lines BndCrcll, BndCrcl2 are specially designed according to the subjective perception of makeup colors, and in particular, these lines are defined only for hues in the a*b* plane that are sensitive to an application of a makeup product. In the example of lipstick-type makeup products, the hues that are sensitive to an application are located approximately on the positive-valued half-plane on a*, namely from yellowish orange hues yORG to bluish violet hues bPRP.

[0084] Furthermore, the chroma boundary lines BndCrcll, BndCrcl2 advantageously depend on the hue of the input coordinate data, in order to take into account the subjective perception of chroma depending on the hue of the respective color. Indeed, for example, orange hues yORG appear "weaker" in terms of chroma than violet hues bPRP. Accordingly, the two chroma boundary lines BndCrcll, BndCrcl2 may have the appearance of a portion of a spiral having a slightly larger radius on positive values ​​on the b* side (yellowish orange hues yORG) and a slightly smaller radius on negative values ​​on the b* side (bluish violet hues bPRP), compared to the spiral radius around the zero value of b* (reddish hues at positive values ​​of a*).

[0085] [Fig.6] illustrates the identification of the position of the input coordinate data 104 relative to at least one hue tone boundary surface BndSrfcl, BndSrfc2 in the L*a*b* space, more particularly in the color family volume ([Fig.3]) including the input coordinate point, used to assign the hue tone subfamily label in step 108 of the method 100.

[0086] In this example, the hue tone subfamily labels include a "warm" hue tone label WRM, a "neutral" hue tone label NTR and a "cool" hue tone label CL and are identified with respect to a first hue tone boundary surface BndSrfcl and a second hue tone boundary surface BndSrfc2, delimiting the space within each respective BRN, ORG, RD, PNK, PRP color family volume.

[0087] The hot label WRM is assigned if the entry point is located on one side of the first hue tone boundary surface BndSrfcl, the neutral label NTR is assigned if the entry point is located between the other side of the first hue tone boundary surface BndSrfcl and one side of the second hue tone boundary surface BndSrfc2, and the cold label CL is assigned if the entry point is located on the other side of the second hue tone boundary surface BndSrfc2.

[0088] Again, the two hue tone boundary surfaces BndSrfcl, BndSrfc2 are specially designed according to the subjective perception of makeup colors, and in particular, the location of these surfaces and the neutral range are determined by the visual results of makeup color experts. The hue tone boundary surfaces BndSrfcl, BndSrfc2 are defined according to the brightness (e.g. from L*=20 to 90) and the chroma (e.g. from C*=0 to the max. level) simultaneously, and their coordinate definition can be recorded in a correspondence table to perform the identification of the relative position of the input coordinate data.

[0089] [Fig.7] illustrates an example of a result at the final step 110 of the classification method 100 described above in relation to FIGS. 1 to 6.

[0090] In the final step 110, the makeup product color is classified according to the assigned labels, advantageously in the illustrated identification card, which illustrates a hierarchical arrangement of colors according to the color family labels on the one hand, the hue tone subfamily labels on the other hand, and then also the brightness subfamily label and the chroma subfamily label.

[0091] On the illustrated identification card, a plurality of colors of makeup products are classified together, each color having been processed according to the method described above in relation to Figures 1 to 6, for the allocation of the respective labels which defines the position of the color on the identification card.

[0092] On the identification card, the colors are arranged accordingly in a table organized by a main row for each color family label "brown", "pink", "orange", "purple", and "red", and a main column for each hue tone subfamily label "cool", "neutral", and "warm". Further, each main row includes a set of subrows for the respective "light", "medium", and "dark" brightness subfamily labels, and each main column includes a set of subcolumns for the respective "low", "intermediate", and "high" chroma subfamily labels.

[0093] Alternatively, the table may be organized without the sets of sub-columns for each chroma sub-family label, replaced by sorting the colors in each brightness sub-row in ascending order of their chroma values, e.g., ascending order from left to right within the respective sub-rows.

[0094] Since the colors are positioned according to their labels on this identification card, namely according to the subjective perceptions they provoke as a makeup product, a color expert can perform an improved analysis of the classified group of makeup colors.

[0095] Indeed, on the one hand, since the classification is carried out automatically by computer, the identification card can be generated several times, for different selections of groups of makeup product colors, in a very rapid manner.

[0096] The method therefore allows the color expert to save a lot of time, compared to the classic technique for this type of classification, carried out manually by visual inspection of miniature color samples, as shown in [Fig.8],

[0097] On the other hand, thanks to the position of the colors corresponding to the subjective perceptions they cause as a makeup product, the identification card is not limited by the color display performance of a given computer screen. Indeed, a computer screen may display the same color for two different data, for example, a low-chroma warm pink may be displayed in an identical manner to a low-chroma cool brown on the computer screen due to its limited performance in terms of "number of colors" compared to the perception sensitivity of the human eye. Such a situation, without the classification on the identification card, leads to an erroneous analysis of these colors.However, the positions of the two colors on the identification chart provide a color expert with information about their practical differences, despite the fact that these two colors are displayed by the same signal on the computer screen.

[0098] [Fig.8] is a photograph of the classic technique of color classification of makeup products, for example to compare products of different brands and franchises of the same company, or existing products owned and products of a competitor.

[0099] Classification is traditionally carried out by color experts with physical samples of the colors under the perception of the human eye, for example with miniature color samples of makeup products which are individually characterized according to subjective perception, and are distributed on a laboratory bench or a whiteboard.

[0100] This conventional technique is obviously very time consuming and is not suitable for, or even impossible to perform, multiple times for many different groups of makeup colors; whereas, as mentioned above, the identification card can be generated as many times as desired, instantly and effortlessly.

[0101] [Fig.9] illustrates an example of a classic computer representation of a group of colors, here in a diagram projected onto the two-dimensional a*b* plane (i.e. ignoring the L* coordinate). Even though the CIELAB a*b* plane was designed to provide a faithful representation of human color perception, such a representation is unusable for analyzing makeup colors. Indeed, even if the identification of the color families "brown", "orange", "pink", "red", "violet" can be partially done visually by an expert in colors on such a diagram, other characterizations are limited to the performance of the computer monitor's color display.

[0102] The above-mentioned problem, where the same color signal is displayed on the computer screen for two different color data (e.g., a low-chroma warm pink can be displayed the same as a low-chroma cool brown), persists and classification according to subjective perception of makeup colors is not possible.

[0103] From another perspective, while the chroma value can be considered as the distance of the color point from the origin, the subjective perception of chroma varies depending on the hue. Therefore, a high-chroma point in a given hue (e.g., a purplish hue) may be inconsistently represented closer to the origin than a lower-chroma point in another hue (e.g., an orange hue), so that the color points may appear mixed and disorganized in the a*b* diagram. Classification according to the subjective perception of makeup colors remains to be done visually and is therefore not feasible due to the limitation of the display performance of the computer monitor.

[0104] [Fig. 10] illustrates a complete computer tool for assisting a color expert, namely a "user", in the development of makeup product colors, benefiting from the classification method and the identification card described above in relation to Figures 1 to 7.

[0105] The tool, which is implemented by a user-controlled computer, may be practically designed as a computer program product comprising instructions which, when the program is executed by a computer, cause the computer to implement the tool in accordance with the present disclosure; or may be practically designed as a computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to implement the tool in accordance with the present disclosure.

[0106] The tool firstly comprises a mapping mode configured to provide the identification map as described above in relation to [Fig.7], for a selected group of makeup colors.

[0107] The user may therefore select a group of makeup product colors from a makeup product color bank, such as for example a bank including makeup colors of internal brands and franchises, and / or makeup colors of competing brands and franchises, to be displayed in mapping mode.

[0108] The user can also select the markets for which the makeup products are intended, or many other options depending on the sectorization commercial color bank.

[0109] In the example of [Fig.7], the identification card therefore classifies the selected color group in the organized table into five main rows for the color family labels brown, pink, orange, violet and red respectively; into three main columns for the subfamily labels cool, neutral and warm hue tone respectively; into three sub-rows for each main row for the subfamily labels light, medium and dark brightness respectively; and into three sub-columns for each main column for the subfamily labels low, intermediate and high chroma respectively.

[0110] Therefore, the mapping mode of the tool provides the user with a visualization of instrumental measurement data in the chart according to the subjective perception of makeup colors. The mapping allows the user to select shades to be visualized on this meaningful identification map, for example a successful competitor's shade range versus an internal shade range, to understand the gaps that the internal brand does not currently cover with shades.

[0111] Figures 11, 12 and 13 illustrate a mode of application of the tool, adapted to digitally simulate the visualization on the face of a range of shades applied, for more specific considerations of the makeup colors selected by the user.

[0112] The application mode is adapted for the user to select a set of one or more makeup product colors from the identification map displayed in the mapping mode, and to select one or more skin tone photography models. The application mode is configured to display simulation images of applications of the selected set of makeup product colors respectively on the at least one skin tone photography model.

[0113] Advantageously, the simulation of applications of the selected set of makeup product colors respectively on the at least one complexion photograph model is carried out by the pseudo-spectral imaging system disclosed in the scientific publication “Liu Z., Xiao K., Pointer M. et al., “Developing a multi-spectral Imaging System using a RGB camera under two illuminations” In: Proceedings of the 28th IS&T Color and Imaging Conference. 28th Color and Imaging Conference, 04-19 Nov 2020, Online. Society for Imaging Science and Technology”; or by the technique disclosed in US patent document No. 8,498,456 B2 issued in the United States (entitled “Method and System for applying cosmetic and / or accessoral enhancements to digital images”); or by the technique disclosed in US patent application document No. 2018 / 0075524 A1 (“Applying virtual makeup products”).

[0114] [Fig. 11] shows the possible options in application mode. In this example, the user has selected approximately nine or more shades from the mapping mode, and the user has selected a skin tone photograph template.

[0115] The simulation images of applications of the selected set of makeup product shades on the selected skin tone photograph model are displayed in a grid that dynamically adapts to the selection. In this example, the grid consists of 3x3 images and can be scrolled up and down to browse through it.

[0116] Two display options may be provided for displaying the simulation images, for example a full face view (see [Fig. 12]) or a close-up view as illustrated in [Fig. 11].

[0117] The shade selection can be changed at any time, by the same process as in mapping mode, for example by using check boxes for each simulation image. In addition, the changed shade selection can be automatically reflected in mapping mode.

[0118] [Fig. 12] shows other possible options in application mode. In this example, the user has selected two shades from the previous mapping mode or application mode, for example from the example application mode in [Fig.l 1], and the user has selected two skin tone photograph models.

[0119] [Fig. 13] shows other possible options in application mode. In this example, the user has selected two shades from the previous mapping mode or application mode, for example from the example application mode in [Fig.l 1], and the user has selected a skin tone photograph template. When only two shades are selected, the application mode offers a side-by-side comparison mode, where moving the cursor left and right instantly compares the two shades.

[0120] [Fig. 14] illustrates a color creation mode of the tool, suitable for digitally creating a new shade and simulating the visualization on the face of the created shade, as a preliminary design step of a new makeup product.

[0121] The color creation mode is adapted for the user to select at least one skin tone photograph model and define parameters for generating a personalized color. The color creation mode is configured to display, in real time, a simulation image of an application of the personalized color on the at least one skin tone photograph model.

[0122] For example, the parameters generating the custom color may include a hue parameter, a saturation parameter, and a brightness parameter. Finish parameters such as “matte,” “satin,” or “glossy” may also be provided.

[0123] In addition, the color creation mode is advantageously adapted for the user to select an existing makeup product color from the displayed identification card, and configured to simultaneously display comparative images of simulations of applications of the customized color and the selected set of makeup product colors, respectively, on the at least one skin tone photograph model. For example, the simultaneous display of the comparative images may be achieved with a sliding cursor as shown in [Fig. 13] or by a grid of simulated images as shown in Figures 11 or 12.

[0124] For example, the user can compare the two shades, by looking at the two shades digitally applied to the face in a side-by-side view, and furthermore with indications of color differences A(E), L*, C*, h(0).

[0125] In summary, the tool described above in relation to Figures 10 to 14 provides a digital system that allows instant analysis of cosmetic color shades and creation of shade ranges, based on an integrated process in interaction with a user according to:

[0126] 1) a data visualization representative of the specific subjective perception makeup color scheme, which allows the user to select shades to view and perform meaningful product and market analysis.

[0127] 2) a digital application of measured nuances on images illustrating models of different skin tones taken using a pseudospectral imaging system, which allows the user to visually assess undertones on models of different skin types by applying shades (e.g., lipstick shades) digitally onto images showing the faces of models with different skin tones, to understand market fit.

[0128] 3) a functionality to accurately design and apply digitally shades on images taken using the aforementioned pseudospectral simulation imaging system, which allows the user to digitally create a shade using an input of brightness, chroma, tint and makeup finish values, which will be rendered in real time on the image of a model's face.

Claims

Claims

1. A computer-implemented method (100) for automatically classifying a color of a makeup product (102), comprising: - providing input coordinate data in L*a*b* space (104) of a makeup product color; - assigning a color family label (106) to the makeup product color according to an identification of a color family volume, among a set of color family volumes in the L*a*b* space (ORG, PNK, RD, PRP, BRN), containing the input coordinate data; the color family volumes (ORG, PNK, RD, PRP, BRN) being designed according to a subjective perception of makeup colors, and the makeup product color being classified according to the assigned labels, - assigning a chroma subfamily label (108) to the makeup product color according to the position of the input coordinate data relative to at least one chroma boundary line (BnCrcll, BndCrcl2) in an a*b* plane of the L*a*b* space; the at least one chroma boundary line (BndCrcll, BndCrcl2) being designed according to a subjective perception of makeup colors, and the makeup product color being classified according to the assigned labels, wherein at least one chroma boundary line (BndCrcll,BndCrcl2) in the a*b* plane of the L*a*b space varies depending on the hue of the input coordinate data, according to the subjective perception of makeup colors.

2. The method of claim 1, wherein the set of color family volumes in L*a*b* space (ORG, PNK, RD, PRP, BRN) is generated from a database (302) comprising a finite number of points in L*a*b* space, each point being labeled with a respective family label according to the subjective perception of makeup colors, and from a mathematical calculation comprising a triangulation (304) generating envelopes encompassing all points of each respective family and an interpolation (306) spreading the envelopes until the respective facing surfaces of neighboring envelopes correspond to each other, the envelopes defining the enclosure of the respective color family volumes.

3. A method according to any one of claims 1 or 2, wherein the color family volumes (BRN, PNK, ORG, PRP, RD) are configured to delineate the colors brown, pink, orange, violet and red in the L*a*b* space according to the subjective perception of makeup colors.

4. A method according to any preceding claim, further comprising: - assigning a brightness subfamily label (108) to the makeup product color according to an identification of the position of the input coordinate data relative to at least one brightness limit value (BndVall, BndVal2) in the brightness coordinate axis L* of the L*a*b* space; the at least one brightness limit value (BndVall, BndVal2) being designed according to a subjective perception of makeup colors, and the makeup product color being classified according to the assigned labels.

5. Method according to claim 4, wherein the at least one brightness limit value (BndVall, BndVal2) in the brightness coordinate axis L* of the L*a*b* space is decreased by one step (C*Stp) for input coordinates having a chroma value greater than a threshold defined according to the subjective perception of makeup colors.

6. The method of any one of claims 4 or 5, wherein the brightness subfamily includes a light label (LGT) assigned if the brightness input coordinate is greater than a first brightness limit value (BndVall), a medium label (MDM) assigned if the brightness input coordinate is between the first brightness limit value (BndVall) and a second brightness limit value (BndVal2), and a dark label (DRK) assigned if the brightness input coordinate is less than the second brightness limit value (BndVal2).

7. A method according to any preceding claim, wherein the chroma subfamily includes a high label (HGH) assigned if the chroma of the input coordinates is greater than a first chroma boundary line (BndCrcll), an intermediate label (INTR) assigned if the chroma of the input coordinates is between the first chroma boundary line (BndCrcll) and a second chroma boundary line (BndCrcl2), and a low label (LW) assigned if the chroma of the input coordinates is less than the second chroma boundary line (BndCrcl2).

8. A method according to any preceding claim, further comprising: - assigning a hue tone subfamily label (108) to the color of the makeup product according to the position of the input coordinate data relative to at least one hue tone boundary surface (BndSrfcl, BndSrfc2) in the respective color family volume in L*a*b* space; the at least one hue tone boundary surface (BndSrfcl, BndSrfc2) being designed according to a subjective perception of makeup colors, and the makeup product color being classified according to the assigned labels.

Images