CN110069716B - Beautiful makeup recommendation method and system and computer-readable storage medium - Google Patents
Beautiful makeup recommendation method and system and computer-readable storage medium Download PDFInfo
- Publication number
- CN110069716B CN110069716B CN201910357049.5A CN201910357049A CN110069716B CN 110069716 B CN110069716 B CN 110069716B CN 201910357049 A CN201910357049 A CN 201910357049A CN 110069716 B CN110069716 B CN 110069716B
- Authority
- CN
- China
- Prior art keywords
- personalized
- face
- decision
- feature
- adjustable
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Abstract
The invention provides a makeup recommendation method, a makeup recommendation system and a computer readable storage medium, wherein the method comprises the following steps: s1, acquiring personalized characteristics of a user; s2, inputting the personalized features into a decision matrix, and generating a recommendation result through operation; the decision matrix comprises a set of decision functions of a plurality of adjustable dimensions of a face single element, and the recommendation result comprises a set of beauty treatment means of a plurality of adjustable dimensions of a face single element. The personalized features of the user are input into a decision matrix formed by a set of decision functions of a plurality of adjustable dimensions comprising a single face element, so that the recommendation result of the set of cosmetic measures of the plurality of adjustable dimensions comprising the single face element is generated through calculation, more detailed cosmetic recommendations can be generated, and personalized guidance can be provided for daily cosmetic activities of the general public practically and effectively.
Description
Technical Field
The invention relates to the technical field of human-computer interaction, in particular to a makeup recommendation method, a makeup recommendation system and a computer-readable storage medium.
Background
In recent years, the pursuit of a beautiful face has led to a vigorous development of the cosmetics and cosmetic industry, resulting in dramatic economic benefits. Makeup is a basic and important part in personal image design, and the makeup learning method adopted by the public mainly comprises watching videos and articles provided by makeup bloggers on the network or speaking and teaching of relatives and friends at home. These methods are not only heavily dependent on the quality of the knowledge source, but also are not targeted to the individual's appearance conditions, so people often need to fumble about a makeup fit for themselves over months or even longer.
For the general public, the current professional and feasible sources of personalized makeup recommendation are mainly offline paid consultation or training service; the internet-related method is to send the photos of the user to professionals such as a makeup blogger, a model designer and a professional makeup artist, and the professionals reply the corresponding word or picture makeup suggestions by the professional makeup artist and the professional makeup artist. However, these methods are still far from normative at present, professional literacy of service personnel is good and irregular, but the charge is generally high, and particularly, offline consultation services are not easy to be burdened by the ordinary people.
Currently, many hypotheses and standards are proposed by scholars regarding the nature and features of highly attractive faces. For example, western rules such as the near classical rule, the golden ratio rule, ancient "santing five eyes" in china, etc.; the academic papers and periodicals published at home and abroad about personal image design and makeup are fewer; in the field of computers, many studies related to makeup have been put on, such as virtual makeup, migration of makeup effects, and the like.
(1) Face photo beautification utilizes a traditional image method and a deep learning method to directly change the physiological characteristics of a face in a photo so as to approach a standard beautiful face.
(2) Virtual makeup is mainly classified into two schemes based on two-dimensional images and three-dimensional images. The user uploads a target face photo, selects a makeup virtual tool (such as foundation, lipstick and eyebrow pencil) provided by the system and then operates the makeup virtual tool, and the system outputs the effect after makeup.
(3) The makeup effect migration is intended to migrate the makeup effect in the reference face image to the target face image, but still preserve the features of the target image.
These studies are only currently in the fields of generation of high-attraction faces, attraction prediction, photo beautification, virtual makeup effect preview and the like, and the current practical application is only the reference function of a cosmetic doctor, so that the improvement of personal images of people in daily life is not substantially facilitated. It is worth mentioning that the existing virtual makeup technology only performs texture processing and image layer superposition effects on the basis of an original image, mainly plays roles of uniform color and enhancing five sense organs, is different from the makeup effect in real life, and cannot provide effective reference for the public. There is still a lack of a recommendation system that can effectively and effectively provide personalized guidance for the daily cosmetic activities of the general public.
Disclosure of Invention
In order to solve the above problems, the present invention provides a makeup recommendation method, a makeup recommendation system, and a computer-readable storage medium.
The invention provides a beauty makeup recommendation method, which comprises the following steps: s1, acquiring personalized characteristics of a user; s2, inputting the personalized features into a decision matrix, and generating a recommendation result through operation; the decision matrix comprises a set of decision functions of a plurality of adjustable dimensions of a face single element, and the recommendation result comprises a set of beauty treatment means of a plurality of adjustable dimensions of a face single element.
In some embodiments, the obtaining personalized features of the user includes: the method comprises the steps of collecting a face image of a user, and carrying out feature recognition on the face image to obtain a first personalized feature. The face image comprises a two-dimensional image or a three-dimensional image; the first characterization feature includes a color feature, a distance feature and an angle feature based on the two-dimensional image, and a depth feature based on the three-dimensional image.
In some embodiments, the obtaining personalized features of the user further comprises: and obtaining the first personalized feature and the second personalized feature which are supplemented in a question-answering mode of man-machine interaction. The second characterization feature includes: favorable vapor bias and cosmetic bias.
Wherein the decision function of the single adjustable dimension of the face single element comprises: a set of personalized features and weights affecting a single tunable dimension of the facial single element.
In some embodiments, further comprising: s3, performing color retouching and presenting on the recommendation result, and receiving a feedback result of the user in the presenting process; the rendering includes adding an interpretation of the recommendation, the interpretation including a text, an annotation, or a practice simulation. And S4, adjusting the decision matrix according to the feedback result, and simultaneously carrying out popularization and propagation.
The present invention also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the cosmetic recommendation method as described above.
The invention also provides a makeup recommendation system, comprising: the characteristic acquisition module is used for acquiring the personalized characteristics of the user; the decision matrix module is used for inputting the personalized features into a decision matrix and generating a recommendation result through operation; the decision matrix comprises a set of decision functions of a plurality of adjustable dimensions of a face single element, and the recommendation result comprises a set of beauty treatment means of a plurality of adjustable dimensions of a face single element.
The invention has the beneficial effects that: the personalized features of the user are input into a decision matrix formed by a set of decision functions of a plurality of adjustable dimensions comprising a single face element, so that the recommendation result of the set of cosmetic measures of the plurality of adjustable dimensions comprising the single face element is generated through calculation, more detailed cosmetic recommendations can be generated, and personalized guidance can be provided for daily cosmetic activities of the general public practically and effectively.
Drawings
Fig. 1 is a flowchart illustrating a cosmetic recommendation method according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of a first characterization feature on a face image according to an embodiment of the present invention.
Fig. 3 is a schematic view illustrating the adjustable dimension splitting of an eyebrow according to an embodiment of the present invention.
Fig. 4 is a schematic view of the adjustable dimension splitting of the lips according to an embodiment of the present invention.
Fig. 5 is a schematic view of the adjustable dimension splitting of an eye in an embodiment of the invention.
Fig. 6 is a block diagram of a makeup recommendation system according to an embodiment of the present invention.
Fig. 7 is a flowchart illustrating functions of the makeup recommendation system from the perspective of the user according to the embodiment of the present invention.
FIGS. 8a-8c are cosmetic reports of embodiments of the present invention.
Detailed Description
The present invention is described in further detail below with reference to specific embodiments and with reference to the attached drawings, it should be emphasized that the following description is only exemplary and is not intended to limit the scope and application of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without any inventive step, are within the scope of the present invention.
The embodiment provides a cosmetic recommendation method, as shown in fig. 1, including the following steps:
and step 100, acquiring personalized features of the user.
For the acquisition of the personalized features of the user, the personalized features of the user can be acquired by acquiring a face image (a two-dimensional image or a three-dimensional image) of the user from multiple angles through a camera and supplementing the acquired face image in a question and answer mode of man-machine interaction. The personalized features comprise first personalized features obtained by acquiring face images and performing feature recognition, and supplementary first personalized features and second personalized features which need to be obtained in a question-answering mode of man-machine interaction.
For the first characterization, may include: color features (such as skin color, hair color, whether black eye is present, etc.), distance features and angle features based on two-dimensional images, and depth features based on three-dimensional images. If the acquired face image is a two-dimensional image, part of the first personalized features (such as depth features) cannot be directly obtained, and the acquired first personalized features may only include: the method comprises the following steps of (1) obtaining a part of first personalized features through a question-answering mode of man-machine interaction by virtue of color features, distance features and angle features; if the acquired face image is a three-dimensional image, various features including, but not limited to, the above can be obtained.
The illustration of the part of the first personalized features on the face image is shown in fig. 2, and comprises distance features (D1-D18, B3, B6) and angle features (a1) based on the two-dimensional image, color features (B1), and depth features (B2, B4, B5) based on the three-dimensional image, which are as follows:
first characterization feature
| Characteristic serial number | Symbol representation | Feature definition |
| 1 | D1 | Longitudinal length of whole face |
| 2 | D2 | Transverse width of whole face |
| 3 | D3 | Longitudinal distance from top of hairline to eyebrow center |
| 4 | D4 | Longitudinal distance from the eyebrow center to the columella nasi |
| 5 | D5 | Longitudinal distance from columella nasi to lowest point of chin |
| 6 | D6 | Width of face at temple height |
| 7 | D7 | Width of face at mouth corner height without expression |
| 8 | D8 | Transverse distance between two inner canthus (concentration degree of five sense organs can be reflected) |
| 9 | D9 | Width of widest part of nose |
| 10 | D10 | Lateral distance from outer canthus to face edge at outer canthus height |
| 11 | D11 | Lateral distance from inner canthus to outer canthus |
| 12 | D12 | Transverse width of eye cleft |
| 13 | D13 | Longitudinal distance from lowest point of middle part of eyebrow to highest point of eye |
| 14 | D14 | Longitudinal height of |
| 15 | D15 | Length in the human body |
| 16 | D16 | Thickness of thickest part of upper lip |
| 17 | D17 | Thickness of thickest part of lower lip |
| 18 | D18 | Longitudinal length of jaw |
| 19 | A1 | Average value of angles formed by connecting lines from two inner canthus to outer canthus and horizontal line |
| 20 | B1 | Color of hair |
| 21 | B2 | Stereo of eyebrow |
| 22 | B3 | Width of double eyelid (0 if single eyelid) |
| 23 | B4 | Height of mountain root |
| 24 | B5 | Sag of temple |
| 25 | B6 | Width in the human being |
The partial first characterization feature described above may be quantitative or qualitative. Such as: width of double eyelids, width in the middle of the human, temple concavity, etc. If the definition of the face image is high, the quantitative feature can be obtained in a feature recognition mode; if the definition of the face image is low, quantitative features are difficult to obtain in a feature recognition mode, and qualitative results of self-perception of a user on certain features can be obtained in a man-machine interaction question-answering mode, such as: the width of the double eyelid is wider/narrower/absent, the width in humans is wider/narrower/absent, the temple is present/absent, etc.
The first characterization feature acquisition step may use a relatively mature method in the prior art, such as face key point detection.
For the second characterization, may include age, gender, skin type, preferred gas bias, preferred make-up bias, etc.; features that the user perceives themselves may also be included, such as: upper/lower half of the face looks good, concentration of five sense organs is high/low, eyebrows are scattered/regular, etc.
And 200, inputting the personalized features into a decision matrix, and generating a recommendation result through operation.
The decision matrix is a set of decision functions comprising a plurality of adjustable dimensions of a face single element, and the recommendation is a set of cosmetic tools comprising a plurality of adjustable dimensions of a face single element.
For the construction of the decision matrix, the specific process is as follows:
1) high-attraction face model based on makeup variable features
FIG. 2 is a high-attractiveness face model, and each first personalized feature has a corresponding reference value. In order to recommend makeup to the face of a user to align with a high-attraction face model, the idea of each large makeup blogger on the network is to perform makeup decoration on a single face element (such as eyebrows, eyes, lips, noses and the like) of the face. Therefore, in this embodiment, by taking this idea as a reference, the correspondence between the features based on the single facial elements and the makeup techniques is first performed, and the factors (personalized features) affecting each single facial element and the makeup techniques corresponding to each factor are analyzed.
As for whether single elements of each face of the human face can influence each other, the improvement space of beauty makeup is not large, so that the defect is mostly compensated, and the compensated degree is not enough to cause large change of the geometric structure of the whole face; therefore, the influence among single elements of each face of the human face can be avoided.
Taking the eyebrows of a single element of the face as an example, the factors (personalized features) affecting the eyebrows and the makeup techniques corresponding to the factors are shown in the following table two.
Second-generation eyebrow influencing factor and beauty technique corresponding to each factor
2) Adjustable dimension splitting of single facial elements
In order to parameterize and code the quantitative standard of the face appeal and the makeup method matched with the quantitative standard to form an executable decision matrix, in the embodiment, all the features of the face are separated into face single elements (such as eyebrows, glasses, lips, noses and the like) which influence the face appeal one by one according to methods such as organs, regions and the like, and each face single element is further split into several adjustable dimensions which do not influence each other from a geometric angle.
In this embodiment, an exhaustive method is adopted for the adjustable dimension splitting process, that is, all adjustable dimensions that geometrically affect the face attraction by changing a single element of the face and can be effectively improved by a cosmetic measure are listed. For example, the thickness of the lips is an adjustable dimension that has an effect on the attractiveness of the face, the value of which can be varied by cosmetic means (varying the thickness of the lips in terms of look and feel using lipstick and foundation products); the width of the lips in the transverse direction is also an adjustable dimension which has an influence on the attractive force of the human face, but the value of the adjustable dimension cannot be changed through a makeup means, so the adjustable dimension is not taken into consideration of a recommendation system; the adjustable dimensions are not mutually influenced, for example, the thicknesses of the upper lip and the lower lip can be respectively changed, and the change of the thickness of the upper lip does not directly cause the change of the thickness of the lower lip.
In one embodiment, the single facial element "eyebrows" is split into 9 mutually unaffected adjustable dimensions: eyebrow position, length, eyebrow distance, thickness eyebrow-eye distance, eyebrow tail height, picking degree, concentration and edge definition. An adjustable dimension splitting schematic diagram of an eyebrow is shown in fig. 3, and the definition of the adjustable dimension is specifically shown in table three below:
adjustable dimension split definition of top three eyebrows
In another embodiment, the face element of "lips" is split into 9 mutually unaffected feature-adjustable dimensions: upper lip thickness, lower lip thickness, upper lip corner shape, lower lip corner shape, lip peak depth, lip peak width, color, texture, edge definition. Fig. 4 shows a schematic diagram of adjustable dimension splitting of lips, and the definition of the adjustable dimension is specifically shown in the following table four:
adjustable dimension split definition of four lips of a watch
| Serial number | (symbol) | Adjustable dimension description/feature definition |
| 1 | Z1 | Thickness of thickest part of upper lip |
| 2 | Z2 | Thickness of thickest part of lower lip |
| 3 | Z3 | Width of lip peak |
| 4 | Z4 | Lip peak depth |
| 5 | Z5 | Colour(s) |
| 6 | Z6 | Texture of |
| 7 | Z7 | Edge definition |
| 8 | Z8 | Upper lip angle shape |
| 9 | Z9 | Lower lip angle shape |
In another embodiment, the face element "eye" is split into 9 mutually unaffected feature-adjustable dimensions: eye line thickness/thinness, eye line drape/teasing, eye line elongation, eye head and eye line, eyelashes, eye shadow shape, eye shadow depth, silkworm sleeping, and pupil beautifying. The adjustable dimension splitting schematic diagram of the eye is shown in fig. 5, and the definition of the adjustable dimension is specifically shown in the following table five:
adjustable dimension split definition for table five eyes
| Serial number | (symbol) | Adjustable dimension description/feature definition |
| 1 | Y1 | Lateral distance of elongation of eye-tail eye line |
| 2 | Y2 | Upward-picking angle at eye-tail eye line stretching position |
| 3 | Y3 | Depth of eye shadow |
| 4 | Y4 | Thickness of eyeliner |
| 5 | Y5 | Beautiful pupil |
| 6 | Y6 | Eye head and eye line |
| 7 | Y7 | Silkworm in bed |
| 8 | Y8 | Shape of eye shadow |
| 9 | Y9 | Eyelash |
3) Constructing a decision matrix
After each element is subjected to adjustable dimension splitting by taking a single facial element as a unit, then according to a makeup method which can be adopted by the single facial element and is used for obtaining a high-attractiveness face, the influence of personalized features on each adjustable dimension, a standard threshold corresponding to each personalized feature and different makeup methods corresponding to the feature value threshold under the upper and lower conditions are sorted out. Different personalized features may correspond to different or even opposite makeup techniques for the same adjustable dimension of the facial element, so different weights need to be given to each personalized feature, and the recommendation system comprehensively analyzes the optimal solution under the limiting condition during calculation.
Taking the single facial element of "eyebrow" as an example, the influence of each personalized feature of different adjustable dimensions of eyebrow on the final recommendation result is shown in the following table six:
decision function with adjustable dimension for table six eyebrows
As can be seen from table six, for the adjustable dimension of "thickness/thickness" of the eyebrow, there are four personalized features (or ratios between the personalized features) that have an effect on the adjustable dimension, where the thresholds of the two features "B2 (the stereoscopy of the eyebrow)" and "D14 (the height of the eye)" limit the thickness of the eyebrow, "D1/D2 (face length/face width)" and "eyebrow distance/eye height" are 2 and 1, respectively, and the weights of the two features are 2/3 and 1/3, respectively, when calculating.
In other embodiments, the decision function for each adjustable dimension for a single face element, the "lips", is as follows:
adjustable dimension decision function for seven-lip watch
| Serial number | Adjustable dimension | Decision function of single adjustable dimension |
| 1 | Degree of lip peak | 0.7 (0.6D 18/D15) + 0.3D 9/D8 to a significant extent |
| 2 | Thickness of upper lip | 0.7 (1.7D15/D18) + 0.3D 8/D9 (upper lip thickness) |
| 3 | Lower lip thickness | (0.6 × D18/D15) ═ lower lip thickness |
| 4 | Upper lip angle shape | (1.4 × D16/D17) ═ degree of sharpness |
| 5 | Lower lip angle shape | (0.7 × D17/D16) ═ degree of sharpness |
| 6 | Colour(s) | Mild, sweet and beautiful |
| 7 | Texture of | Pure and thick matte |
| 8 | Definition of | Definition of |
In another embodiment, for a single face element, the "eyes", the decision function for each adjustable dimension is shown in table eight below:
TABLE eight eye decision function with adjustable dimensionality
| Serial number | Adjustable dimension | Decision function |
| 1 | Thickness/fineness of eye line | B3 (limit); 1.2 × D10/D8 ═ accentuating eyesTail |
| 2 | Eye line plumbing/picking | A1/10 degree (self-alignment) |
| 3 | Degree of elongation of eyeliner | D12 (limit); 1.2 × D10/D8-degree of elongation |
| 4 | Eye head and eye line | D11/D8 ═ eye head and eye line |
| 5 | Eyelash | 0.8 × D13/D14 ═ mascara/false eyelash necessity |
| 6 | Shape of eye shadow | 1.2 × D10/B2+ D14/0.8 × D13 ═ degree of elongation |
| 7 | Depth of eye shadow | Orbital depth (limit) 0.8 × D13/D14 |
| 8 | Silkworm in bed | Black eye depth (limiting) necessary degree of lying silkworms |
| 9 | Beautiful pupil | Necessary degree of pupil beautification |
According to the influence of each personalized feature of different adjustable dimensions on the final recommendation result, the influence conditions are sorted into decision functions of the adjustable dimensions capable of outputting a single value by using mathematical thinking, and then the decision functions of each adjustable dimension are integrated to form a decision matrix of a single element of the face.
And correspondingly inputting the first personalized features acquired in the step 100 into a decision matrix of a single element of each face, and generating a recommendation result through operation.
In addition, in step 100, in addition to the first personalized feature, a second personalized feature of the user is obtained. In order to fully consider the individual beauty requirements of the user, two aspects of consideration such as 'gas quality bias' and 'cosmetic feel bias' are added to each adjustable dimension of each single face element in the construction process of the recommendation result. And for different adjustable dimensions, different cosmetic manipulations have different influences on the aspects of 'quality of breath' and 'attractiveness', so that different weights are given to the two subjective items.
For a single element of the face of the eyebrow, different scores are set to represent different preference and cosmetic deviation of the user, in the ninth table, the scores are set to be-2, -1, 0, 1 and 2, the lower the score is, the more the user is biased to a soft and natural gas style, and the higher the score is, the more the user is biased to a delicate gas style.
Vapor/cosmetic feel value for each adjustable dimension in the top nine eyebrows
The following table ten is a weight function of the gas-quality weight for each adjustable dimension in the eyebrow, and after a decision matrix is used for calculating a recommended value of a certain adjustable dimension for the eyebrow, the 'gas-quality addition' weight value of the adjustable dimension is multiplied by a result to obtain a final makeup recommended result of the adjustable dimension, wherein the adjustable dimension comprehensively considers the subjective preference of the user.
Vapor weight for each adjustable dimension of the table brow
Based on the makeup recommendation method, the personalized features of the user are input into the decision matrix formed by the set of decision functions of the multiple adjustable dimensions comprising the single face element, so that the recommendation result of the set of makeup means of the multiple adjustable dimensions comprising the single face element is generated through calculation, more detailed makeup recommendation can be generated, and personalized guidance can be provided for daily makeup activities of the general public practically and effectively.
The beauty makeup recommendation method further comprises the following steps:
And step 400, adjusting the decision matrix according to the feedback result, and simultaneously carrying out popularization and propagation.
The present embodiment also provides a system, as shown in fig. 6, including: the system comprises an individualized characteristic acquisition module, a decision matrix module, a color matching module, a user feedback module and a popularization and propagation module.
Acquiring personalized features of a user by means of face image acquisition and man-machine interaction question answering in a personalized feature acquisition module, inputting the personalized features into a decision matrix and generating a recommendation result through operation; the decision matrix comprises a set of decision functions of a plurality of adjustable dimensions of a face single element, and the recommendation result comprises a set of beauty treatment means of the plurality of adjustable dimensions of the face single element. And then, the recommendation result is retouched (including adding explanations to the recommendation result, such as pictures and texts, notes or practice simulation diagrams) through a retouching module and presented to the user, the user submits feedback through a feedback module and carries out popularization and propagation through a popularization and propagation module, more benefits are obtained, the popularization of the product is facilitated, and more users use the product to input data.
For the user perspective, the functional flow charts of the makeup recommendation system are shown in fig. 7. For the cosmetic report after touch-up, a partial schematic is shown in FIGS. 8a-8 c.
Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but the former is a better implementation mode in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method of the embodiments of the present invention.
The foregoing is a more detailed description of the invention in connection with specific/preferred embodiments and is not intended to limit the practice of the invention to those descriptions. It will be apparent to those skilled in the art that various substitutions and modifications can be made to the described embodiments without departing from the spirit of the invention, and these substitutions and modifications should be considered to fall within the scope of the invention.
Claims (9)
1. A cosmetic recommendation method is characterized by comprising the following steps:
s1, acquiring personalized characteristics of a user;
s2, inputting the personalized features into a decision matrix, and generating a recommendation result through operation; the decision matrix comprises a set of decision functions of a plurality of adjustable dimensions of a face single element, and the recommendation result comprises a set of beauty treatment means of a plurality of adjustable dimensions of the face single element;
the decision function of the adjustable dimension comprises: a set of personalized features and weights affecting a single tunable dimension of the facial single element;
the construction of the decision matrix comprises the following steps: analyzing personalized features influencing each single facial element and a makeup technique corresponding to each personalized feature based on a high-attraction face model with makeup variable features; adjustable dimension splitting of a single facial element; constructing a decision matrix: according to the influence of each personalized feature of different adjustable dimensions on the final recommendation result, the influence conditions are sorted into decision functions of the adjustable dimensions capable of outputting a single value by using mathematical thinking, and then the decision functions of each adjustable dimension are integrated to form a decision matrix of a single element of the face.
2. The cosmetic recommendation method of claim 1, wherein the obtaining personalized features of the user comprises: the method comprises the steps of collecting a face image of a user, and carrying out feature recognition on the face image to obtain a first personalized feature.
3. The cosmetic recommendation method according to claim 2, wherein the face image comprises a two-dimensional image or a three-dimensional image; the first characterization feature includes a color feature, a distance feature and an angle feature based on the two-dimensional image, and a depth feature based on the three-dimensional image.
4. The cosmetic recommendation method of claim 2, wherein the obtaining personalized features of the user further comprises: and obtaining the first personalized feature and the second personalized feature which are supplemented in a question-answering mode of man-machine interaction.
5. The cosmetic recommendation method of claim 4, wherein the second characterization feature comprises: favorable vapor bias and cosmetic bias.
6. The cosmetic recommendation method of claim 1, further comprising:
s3, performing color retouching and presenting on the recommendation result, and receiving a feedback result of the user in the presenting process; the rendering includes adding an interpretation of the recommendation, the interpretation including a text, an annotation, or a practice simulation.
7. The cosmetic recommendation method of claim 6, further comprising:
and S4, adjusting the decision matrix according to the feedback result, and simultaneously carrying out popularization and propagation.
8. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the cosmetic recommendation method according to any one of claims 1 to 7.
9. A cosmetic recommendation system, comprising:
the personalized feature acquisition module is used for acquiring personalized features of the user;
the decision matrix module is used for inputting the personalized features into a decision matrix and generating a recommendation result through operation; the decision matrix comprises a set of decision functions of a plurality of adjustable dimensions of a face single element, and the recommendation result comprises a set of beauty treatment means of a plurality of adjustable dimensions of the face single element;
the decision function of the adjustable dimension comprises: a set of personalized features and weights affecting a single tunable dimension of the facial single element;
the construction of the decision matrix module comprises the following steps: analyzing personalized features influencing each single facial element and a makeup technique corresponding to each personalized feature based on a high-attraction face model with makeup variable features; adjustable dimension splitting of a single facial element; constructing a decision matrix: according to the influence of each personalized feature of different adjustable dimensions on the final recommendation result, the influence conditions are sorted into decision functions of the adjustable dimensions capable of outputting a single value by using mathematical thinking, and then the decision functions of each adjustable dimension are integrated to form a decision matrix of a single element of the face.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910357049.5A CN110069716B (en) | 2019-04-29 | 2019-04-29 | Beautiful makeup recommendation method and system and computer-readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910357049.5A CN110069716B (en) | 2019-04-29 | 2019-04-29 | Beautiful makeup recommendation method and system and computer-readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110069716A CN110069716A (en) | 2019-07-30 |
| CN110069716B true CN110069716B (en) | 2022-03-18 |
Family
ID=67369571
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910357049.5A Active CN110069716B (en) | 2019-04-29 | 2019-04-29 | Beautiful makeup recommendation method and system and computer-readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110069716B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111553219A (en) * | 2020-04-21 | 2020-08-18 | 海信集团有限公司 | Data processing method and intelligent equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107153805A (en) * | 2016-03-02 | 2017-09-12 | 北京美到家科技有限公司 | Customize makeups servicing unit and method |
| CN108229415A (en) * | 2018-01-17 | 2018-06-29 | 广东欧珀移动通信有限公司 | Information recommendation method, device, electronic equipment and computer readable storage medium |
| KR20180110842A (en) * | 2017-03-30 | 2018-10-11 | 최진은 | Customized semi-permanent make-up recommendation system based on virtual experience and its service method |
| CN108682050A (en) * | 2018-08-16 | 2018-10-19 | Oppo广东移动通信有限公司 | U.S. face method and apparatus based on threedimensional model |
| CN108920490A (en) * | 2018-05-14 | 2018-11-30 | 京东方科技集团股份有限公司 | Assist implementation method, device, electronic equipment and the storage medium of makeup |
| CN109671142A (en) * | 2018-11-23 | 2019-04-23 | 南京图玩智能科技有限公司 | A kind of intelligence makeups method and intelligent makeups mirror |
-
2019
- 2019-04-29 CN CN201910357049.5A patent/CN110069716B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107153805A (en) * | 2016-03-02 | 2017-09-12 | 北京美到家科技有限公司 | Customize makeups servicing unit and method |
| KR20180110842A (en) * | 2017-03-30 | 2018-10-11 | 최진은 | Customized semi-permanent make-up recommendation system based on virtual experience and its service method |
| CN108229415A (en) * | 2018-01-17 | 2018-06-29 | 广东欧珀移动通信有限公司 | Information recommendation method, device, electronic equipment and computer readable storage medium |
| CN108920490A (en) * | 2018-05-14 | 2018-11-30 | 京东方科技集团股份有限公司 | Assist implementation method, device, electronic equipment and the storage medium of makeup |
| CN108682050A (en) * | 2018-08-16 | 2018-10-19 | Oppo广东移动通信有限公司 | U.S. face method and apparatus based on threedimensional model |
| CN109671142A (en) * | 2018-11-23 | 2019-04-23 | 南京图玩智能科技有限公司 | A kind of intelligence makeups method and intelligent makeups mirror |
Non-Patent Citations (1)
| Title |
|---|
| 东方女性人脸妆容推荐算法研究;李阳帆;《中国优秀硕士学位论文全文数据库 信息科技辑》;20190115;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110069716A (en) | 2019-07-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10799010B2 (en) | Makeup application assist device and makeup application assist method | |
| KR101259662B1 (en) | Face classifying method face classifying device classification map face classifying program recording medium where this program is recorded | |
| WO2014119242A1 (en) | Makeup application assistance device, makeup application assistance method, and makeup application assistance program | |
| Kagian et al. | A humanlike predictor of facial attractiveness | |
| Zhang et al. | Computer models for facial beauty analysis | |
| Kagian et al. | A machine learning predictor of facial attractiveness revealing human-like psychophysical biases | |
| JP2005038375A (en) | Eye configuration classifying method and configuration classifying map and eye make-up method | |
| Gunes | A survey of perception and computation of human beauty | |
| Park et al. | An automatic virtual makeup scheme based on personal color analysis | |
| CN110069716B (en) | Beautiful makeup recommendation method and system and computer-readable storage medium | |
| Bottino et al. | The analysis of facial beauty: an emerging area of research in pattern analysis | |
| Guo | Digital anti-aging in face images | |
| JP5095182B2 (en) | Face classification device, face classification program, and recording medium on which the program is recorded | |
| Graham et al. | Representation and aesthetics of the human face in portraiture | |
| Lo et al. | Evaluating the effect of interactions between appearance-related product designs and facial characteristics on social affectivity | |
| Varshovi | Facial makeup detection using HSV color space and texture analysis | |
| US20210279933A1 (en) | Method and system to provide a computer-modified visualization of the desired face of a person | |
| Yang et al. | Example-based caricature generation with exaggeration control | |
| JP4893968B2 (en) | How to compose face images | |
| Feng et al. | Quantifying the makeup effect in female faces and its applications for age estimation | |
| JP5650012B2 (en) | Facial image processing method, beauty counseling method, and facial image processing apparatus | |
| Xu et al. | Synthesis of facial caricature using eigenspaces | |
| Li et al. | Large-pose facial makeup transfer based on generative adversarial network combined face alignment and face parsing | |
| CN104715505A (en) | Three-dimensional head portrait generating system and generating device and generating method thereof | |
| CN204791190U (en) | Three -dimensional head portrait generation system and device thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |