CN109255328A - User's makings determines method and device - Google Patents
User's makings determines method and device Download PDFInfo
- Publication number
- CN109255328A CN109255328A CN201811045713.4A CN201811045713A CN109255328A CN 109255328 A CN109255328 A CN 109255328A CN 201811045713 A CN201811045713 A CN 201811045713A CN 109255328 A CN109255328 A CN 109255328A
- Authority
- CN
- China
- Prior art keywords
- point
- characteristic
- eye
- nose
- face
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of user's makings to determine method and device, is related to facial information process field.User's makings provided by the invention determines method, when realizing, obtains the two-dimentional face image of user first;Then, two-dimentional face image is input to feature identification model, to determine facial feature points;Then, face characteristic information is calculated according to facial feature points;Finally, determining the makings of user according to face characteristic information.By introducing feature identification model, so that determining that the mode of user's facial feature points is more standardized, so that determining that the mode of user's makings is more reasonable.
Description
Technical field
The present invention relates to facial information process fields, determine method and device in particular to user's makings.
Background technique
With the improvement of the quality of life, people increasingly pay attention to appearance.The people bad for appearance makings, it will usually select
The mode of face's shaping is selected to improve the appearance makings of itself.
In face's shaping field and related fields, cosmetic surgeons would generally be according to the actual conditions of user come for user
Determine corresponding face's shaping policy, some for example padded some position on the face of face's shaping policy changes some position on the face
Form.After the completion of shaping, user can judge whether shaping reaches expected mesh by the photo before and after comparison shaping
's.
Makings shaping is novel shaping direction, i.e., personal makings is improved by way of shaping.
Summary of the invention
The purpose of the present invention is to provide a kind of user's makings to determine method.
In a first aspect, the embodiment of the invention provides a kind of user's makings to determine method, comprising:
Obtain the two-dimentional face image of user;
Two-dimentional face image is input to feature identification model, to determine facial feature points;
Face characteristic information is calculated according to facial feature points;
The makings of user is determined according to face characteristic information.
Second aspect, the embodiment of the invention provides a kind of user's makings determining devices, comprising:
Module is obtained, for obtaining the two-dimentional face image of user;
Input module, for two-dimentional face image to be input to feature identification model, to determine facial feature points;
Computing module, for calculating face characteristic information according to facial feature points;
Determining module, for determining the makings of user according to face characteristic information.
User's makings provided in an embodiment of the present invention determines method, when realizing, obtains the two-dimentional face of user first
Image;Then, two-dimentional face image is input to feature identification model, to determine facial feature points;Then, according to face feature
Point calculates face characteristic information;Finally, determining the makings of user according to face characteristic information.Mould is identified by introducing feature
Type, so that determining that the mode of user's facial feature points is more standardized, so that determining that the mode of user's makings is more reasonable.
To enable the above objects, features and advantages of the present invention to be clearer and more comprehensible, preferred embodiment is cited below particularly, and cooperate
Appended attached drawing, is described in detail below.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 shows the basic flow chart that user's makings provided by the embodiment of the present invention determines method;
Fig. 2 shows user's makings provided by the embodiment of the present invention to determine in method, and the positive face face image of two dimension shows
It is intended to;
Fig. 3 shows user's makings provided by the embodiment of the present invention and determines in method, the schematic diagram of 90 ° of side face images;
Fig. 4 shows user's makings provided by the embodiment of the present invention and determines in method, the schematic diagram of 45 ° of side face images;
Fig. 5 shows user's makings provided by the embodiment of the present invention and determines in method, the signal of face corresponding to example 1
Figure;
Fig. 6 shows user's makings provided by the embodiment of the present invention and determines in method, the signal of face corresponding to example 2
Figure;
Fig. 7 shows user's makings provided by the embodiment of the present invention and determines in method, the signal of face corresponding to example 3
Figure;
Fig. 8 shows user's makings provided by the embodiment of the present invention and determines in method, the signal of face corresponding to example 4
Figure;
Fig. 9 shows user's makings provided by the embodiment of the present invention and determines in method, the signal of face corresponding to example 5
Figure;
Figure 10 shows user's makings provided by the embodiment of the present invention and determines in method, and face corresponding to example 6 shows
It is intended to;
Figure 11 shows user's makings provided by the embodiment of the present invention and determines in method, and face corresponding to example 7 shows
It is intended to.
Specific embodiment
Below in conjunction with attached drawing in the embodiment of the present invention, technical solution in the embodiment of the present invention carries out clear, complete
Ground description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Usually exist
The component of the embodiment of the present invention described and illustrated in attached drawing can be arranged and be designed with a variety of different configurations herein.Cause
This, is not intended to limit claimed invention to the detailed description of the embodiment of the present invention provided in the accompanying drawings below
Range, but it is merely representative of selected embodiment of the invention.Based on the embodiment of the present invention, those skilled in the art are not doing
Every other embodiment obtained under the premise of creative work out, shall fall within the protection scope of the present invention.
In the related technology, the deterministic process of user's makings usually largely will receive the influence of user's subjectivity, this makes
It is not objective enough for the judgement of user's makings to obtain.
In view of this, this application provides a kind of user's makings to determine method, as shown in Figure 1, including the following steps:
S101 obtains the two-dimentional face image of user;
Two-dimentional face image is input to feature identification model, to determine facial feature points by S102;
S103 calculates face characteristic information according to facial feature points;
S104 determines the makings of user according to face characteristic information.
Wherein, face's two dimensional image of user can be shot by video camera after directly obtain, can also be
The three-dimensional face image for first obtaining user then maps out corresponding two-dimentional face image, the two dimension from three-dimensional face image
Face image can be face image, 90 ° of side face images (image at the visual angle that the visual angle angle with face image is 90), can also
To be 45 ° of side face images, certainly, two-dimentional face image is also possible to the side face image of other angles, but the two of these three angles
Tie up face image (face image, 90 ° of side face images and 45 ° of side face images) more targetedly.
Two-dimentional face image generated in step S101 can be the image at a visual angle, and (two-dimentional face image is one
A image), the image for being also possible to different perspectives (in such as two-dimentional face image while including face image and 90 ° of side face figures
Picture).
In step S102, two-dimentional face image is inputted into value tag identification model, feature identification model is exported
Facial feature points, specifically, the specific coordinate that can be facial feature points of feature identification model output.
Then, it in step S103, can directly be carried out using the specific coordinate of facial feature points as face characteristic information defeated
Out, for example, left labial angle point and right labial angle point can reflect the width of mouth in turn can be directly by left labial angle point and right labial angle
Point is exported as face characteristic information;It is also possible to first calculate mouth width according to left labial angle point and right labial angle point, and
Mouth width is exported.
When specific implementation, the facial feature points determined in step S102 may be it is multiple, in that case,
The specific implementation process of step S103 can be the facial feature points conduct selected and can exported from multiple facial feature points
Face characteristic information.
After face characteristic information has been determined, the feature of user can be determined with direct basis face characteristic information, than
Such as, the shape of the size of eyes of user, eyes can reflect the makings situation of user.It should be noted that face characteristic
There are many kinds of information is actually, in general, in order to which the makings of user is determined more accurately, should recognize it is enough
Face characteristic information and then the makings of determining user are (after being typically required for 5-10 or more face characteristic informations,
The makings of sufficiently exact determining user).
Such as explanation hereinbefore, two-dimentional face image is directly by 2D video camera to people respectively there are two types of acquisition modes
After face is shot, get;And the three-dimensional face image of user is first obtained, then mapped from three-dimensional face image
Corresponding two-dimentional face image out.If provided herein by the way of the three-dimensional face image for first obtaining user
In method, step S101 includes:
Step 1011, three-dimensional face image is obtained;Three-dimensional face image is the face using three-dimensional scanner to user
It is scanned;
Step 1012, the mapping of specified angle is carried out, to three-dimensional face image to determine two-dimentional face image;The two dimension
Face image includes any one or more image below: face image, 90 ° of side face images and 45 ° of side face images.
In step 1011, the human face scanning carried out by special 3D human face scanning device to user can be.3D face
Scanning means can be hand-held, be also possible to console mode.It is fixed that scanning mode can be scanning means, head part's rotation
It is scanned;It is also possible to that head part is fixed, the camera in scanning means automatically turns along some track around face
It is dynamic to be scanned.For guarantee scanning accuracy, it is preferable to use console mode 3D human face scanning device, camera in scanning means from
Dynamic rotates the mode being scanned around face along some track, completes the work of 3D scanning.It is excellent on 3D human face scanning device
Choosing is provided with LED illumination device and crown fixed cell, and the main function of the LED illumination device is when carrying out 3D shooting
It waits, illumination is provided;Crown fixed cell when in use, can be fixed on the head of user, so that the head of user will not
It is subjected to displacement.
In step 1012, mainly the three-dimensional face image in step 1011 is mapped according to specified angle, with
Determining two-dimentional face image corresponding to the specified angle, the angle of mapping can be determined according to the actual conditions of user, but
It should be noted that feature identification model is typically only capable to be known for the two-dimentional face image of some angle in step S102
Not, therefore, when executing step 1012, it is necessary first to determine feature identification model can identify in step S102 two
The angle of face image is tieed up, and determines the specific value of specified angle in step 1012 according to the angle.
Using from three-dimensional face image map two dimensional image by the way of, advantage be 3D scanning during to face
Angle does not have excessively high requirement, and the 3-D image after scanning can be automatically adjusted the angle of face according to the parameter of setting,
The two dimensional image mapped after adjustment can be used directly.And two dimensional image is shot by video camera, then it needs manually to ajust people in advance
The angle of face, is shot.The angle requirement that face is ajusted is stringent, not easy to operate, and the two dimensional image of acquisition is easy to appear data
Deviation.
Specifically, determining facial feature points are either one or more in step S102.Under normal conditions,
These facial feature points can be divided into different groups according to the difference on position on the face, and facial feature points include following
Any one or more feature groups: Zheng Lian feature group, 90 ° of side face feature groups and 45 ° of side face feature groups;
Meanwhile Zheng Lian feature group includes any one or more characteristic point groups below: outer profile point group looks straight
Bu Dian group, nose point group and mouth point group;
90 ° of side face feature groups include any one or more characteristic point groups below:
Side profile point group, branch hole Bu Dian group and point group of ear;
45 ° of side face feature groups include any one or more characteristic point groups below:
Oblique profile point group;
Outer profile point group includes following any one or more characteristic points:
Metopion, left frontal eminence horizontal extension point, right frontal eminence horizontal extension point, left temples salient point, right temples salient point, left temples
Concave point, right temples concave point, left cheekbone high point, right cheekbone high point, left cheekbone lower edge point, right cheekbone lower edge point, left jaw angle point, right jaw
Angle point and chin bottom point;
Positive eye point group group includes following any one or more characteristic points:
Upper left brows starting point, upper right brows starting point, left eyebrow tail point, right eyebrow tail point, upper left eye socket vertex, upper right eye socket top
Point, left eye endocanthion point, right eye endocanthion point, upper left margo palpebrae vertex, upper right margo palpebrae vertex, left eye outer canthus point, right eye outer canthus point, lower-left
Margo palpebrae bottom point and bottom right margo palpebrae bottom point;
Nose point group includes following any one or more characteristic points:
Space between the eyebrows point, left nasal base breadth point, right nasal base breadth point, left nose hole vertex, right nostril vertex and muffle angle point;
Mouth point group includes following any one or more characteristic points:
Lip valley point, left labial angle point, right labial angle point, lower lip bottom point, lip pearl midpoint and lower interior labrale;
Side profile point group includes following any one or more characteristic points:
The excessive point of volume, side volume end point, side space between the eyebrows point, side nose mountain root point, side bridge of the nose starting point, the side bridge of the nose under side metopion, side
High point, side nose starting point, side nose vertex, side nostril vertex, side nose pillar midpoint, side muffle angle point, midpoint in people from side, on side
Lip inflection point, side chin labial groove concave point, the excessive point of side chin, side chin lower edge point, side neck jaw inflection point under vertex, side lower lip;
Branch hole Bu Dian group includes following any one or more characteristic points:
Branch hole socket of the eye vertex, side upper eyelid outer rim vertex, side lower eyelid outer rim bottom point, side nose basal point and side corners of the mouth point;
Point group of ear includes following any one or more characteristic points:
Side inferior crura of antihelix edge point and bottom point of hanging down of picking up the ears;
Oblique profile point group includes following any one or more characteristic points:
Oblique upper zygion, oblique cheekbone inflection point, oblique cheekbone transition point, oblique apple flesh high point, tiltedly interior cheek starting point, tiltedly interior cheek
Point, tiltedly interior cheek end point and oblique chin starting point.
When specific implementation, facial feature points can be outer profile point group, side profile point group, oblique profile point group, just
In eye point group, branch hole Bu Dian group, nose point group, mouth point group and point group of ear, any one or more groups
Point in group.
In order to more accurately illustrate the position of above-mentioned each characteristic point, illustrate each characteristic point in the form of a list below
In specific location on the face, as shown in table 1- table 3:
Table 1: the positive face face image (as shown in Figure 2) of two dimension
2:90 ° of side face image (as shown in Figure 3) of table
3:45 ° of side face image (as shown in Figure 4) of table
When specific implementation, the characteristic point in table 1 is according to two-dimentional positive face face image (one in two-dimentional face image
Kind) determine;Characteristic point in table 2 is determined according to 90 ° of side face images (one of two-dimentional face image);In table 3
Characteristic point be according to 1-89 ° of side face image (one of two-dimentional face image, such as 45 ° of side face images) determine.
It can be seen from the above content that the image of different angle is for determining different characteristic points, such as two-dimentional positive face
Face image is for determining the characteristic point in outer profile point group, the characteristic point in positive eye point group group, in nose point group
Characteristic point and mouth point group in characteristic point;90 ° of side face images are for determining the characteristic point in side profile point group, side
The characteristic point in characteristic point, point group of ear in eye point group;45 ° of side face images are for determining in oblique profile point group
Characteristic point.
Namely step S102 can be executed according to any one or more following mode:
First way: two-dimentional positive face face image is input to feature identification model, to determine Zheng Lian feature group, just
Face feature group includes the characteristic point in following any one or more groups: outer profile point group, positive eye point group group, nose
Point group and mouth point group;
The second way: being input to feature identification model for 90 ° of side face images, with determine 90 ° of side face feature groups, 90 °
Side face feature group includes the characteristic point in following any one or more groups: side profile point group, branch hole Bu Dian group, ear
Bu Dian group;
The third mode: being input to feature identification model for 45 ° of side face images, with determine 45 ° of side face feature groups, 45 °
Side face feature group includes the characteristic point in following group:
Oblique profile point group.
By the characteristic point in above-mentioned table 1-3, user can voluntarily select which characteristic point to calculate face characteristic using
Information, that is, the scoring of certain face areas is calculated, specifically,
Face characteristic information includes any one or more characteristic informations below:
Positive face characteristic information, 90 ° of side face characteristic informations and 45 ° of side face characteristic informations;
Positive face characteristic information includes below any one or more:
Face's length-width ratio, upper presiding judge's degree, face amount is wide, temporo is wide, bizygomatic, outer cheek plumpness, jaw are wide, lower presiding judge's degree, lower chin
Length, symmetry, face Evaluation on distribution value, eyebrow tail length, eyebrow raise angle, glabella away from, looks spacing, eye is long, palpebral fissure size, eye are big
Small, eye raises that angle, eye spacing, Zhongting length, wing of nose width, the exposed degree in nostril, length, lip separated time, lip raise angle, lip closure in people
Degree, mouth size, upper lip thin and thick, lower lip thin and thick, jaw structure evaluation of estimate;
90 ° of side face characteristic informations include below any one or more:
Volume head trim, forehead plumpness, space between the eyebrows height, eye convexity, nasion height, nasion form, contour of nasal bridge, nose are high
Degree, nose substrate height, nasal labial angle, upper lip convex flat, convex flat, the lower chin set-back of lower lip, lower chin head trim, lower jaw edge clarity, jawbone knot
Structure evaluation of estimate;
45 ° of side face characteristic informations include below any one or more:
Cheekbone height, apple flesh plumpness, interior cheek plumpness.
In turn, step S103 according to facial feature points calculate face characteristic information can according to it is following any one or more
Mode executes:
According to the characteristic point in Zheng Lian feature group, positive face characteristic information is calculated;
According to the characteristic point in 90 ° of side face feature groups, 90 ° of side face characteristic informations are calculated;
According to the characteristic point in 45 ° of side face feature groups, 45 ° of side face characteristic informations are calculated.
The characteristic point that each positive face characteristic information uses is calculated specifically, being set forth below out, but it should know,
When actually calculating, it can be calculated, can also be carried out according to the actual situation fully according to following disclosed characteristic point
Fine tuning, but should at least use a part of corresponding characteristic point.
Calculate the characteristic point of face's length-width ratio:
Metopion, chin bottom point, left cheekbone high point, right cheekbone high point;
The characteristic point of presiding judge's degree in calculating:
Metopion, space between the eyebrows point, muffle angle point;
Calculate the wide characteristic point of face amount:
Left temples salient point, right temples salient point, left cheekbone high point, right cheekbone high point;
Calculate the wide characteristic point of temporo:
Left temples concave point, right temples concave point, left cheekbone high point, right cheekbone high point;
Calculate the characteristic point of bizygomatic:
Left temples concave point, right temples concave point, left cheekbone high point, right cheekbone high point, left cheekbone lower edge point, right cheekbone lower edge
Point;
Calculate the characteristic point of outer cheek plumpness:
Left cheekbone high point, right cheekbone high point, left cheekbone lower edge point, right cheekbone lower edge point, left jaw angle point, right jaw angle point;
Calculate the wide characteristic point of jaw:
Left cheekbone high point, right cheekbone high point, left jaw angle point, right jaw angle point.
Calculate the characteristic point of lower presiding judge's degree:
Space between the eyebrows point, muffle angle point, chin bottom point.
Calculate the characteristic point of lower chin length:
Space between the eyebrows point, muffle angle point, lower lip bottom point, chin bottom point.
Calculate the characteristic point of symmetry:
Space between the eyebrows point, muffle angle point, left labial angle point, right labial angle point.
Calculate the characteristic point of face Evaluation on distribution value:
Metopion, left frontal eminence horizontal extension point, right frontal eminence horizontal extension point, left cheekbone high point, right cheekbone high point, left jaw angle
Point, right jaw angle point, chin bottom point, left eye outer canthus point, right eye outer canthus point, left labial angle point, right labial angle point.
Calculate the characteristic point of eyebrow tail length:
Upper left brows starting point, upper right brows starting point, left eyebrow tail point, right eyebrow tail point, left cheekbone high point, right cheekbone high point.
Calculate the characteristic point that eyebrow raises angle:
Upper left brows starting point, upper right brows starting point, left eyebrow tail point, right eyebrow tail point.
Calculate glabella away from characteristic point:
Upper left brows starting point, upper right brows starting point, left cheekbone high point, right cheekbone high point.
Calculate the characteristic point of looks spacing:
Space between the eyebrows point, muffle angle point, upper left eye socket vertex, upper right eye socket vertex, upper left margo palpebrae vertex, upper right margo palpebrae vertex.
Calculate the long characteristic point of eye:
Left eye endocanthion point, right eye endocanthion point, left eye outer canthus point, right eye outer canthus point, left cheekbone high point, right cheekbone high point.
Calculate the characteristic point of palpebral fissure size:
Upper left margo palpebrae vertex, upper right margo palpebrae vertex, lower-left margo palpebrae bottom point, bottom right margo palpebrae bottom point, left cheekbone high point, right cheekbone
High point
Calculate the characteristic point of eye size:
Metopion, left frontal eminence horizontal extension point, right frontal eminence horizontal extension point, left cheekbone high point, right cheekbone high point, left jaw angle
Point, right jaw angle point, chin bottom point, left eye endocanthion point, right eye endocanthion point, upper left margo palpebrae vertex, upper right margo palpebrae vertex, left eye outer canthus
Point, right eye outer canthus point, lower-left margo palpebrae bottom point, bottom right margo palpebrae bottom point.
Calculate the characteristic point that eye raises angle:
Left eye endocanthion point, right eye endocanthion point, left eye outer canthus point, right eye outer canthus point.
Calculate the characteristic point of eye spacing:
Left eye outer canthus point, right eye outer canthus point, left cheekbone high point, right cheekbone high point.
Calculate the characteristic point of Zhongting length:
Metopion, space between the eyebrows point, muffle angle point, chin bottom point.
Calculate the characteristic point of wing of nose width:
Left cheekbone high point, right cheekbone high point, left nasal base breadth point, right nasal base breadth point.
Calculate the characteristic point in nostril:
Left nose hole vertex, right nostril vertex, muffle angle point.
Calculate the characteristic point of length in people:
Space between the eyebrows point, muffle angle point, lip valley point.
Calculate the characteristic point of lip separated time:
Muffle angle point, lip pearl midpoint, chin bottom point.
Calculate the characteristic point that lip raises angle:
Lip pearl midpoint, left labial angle point, right labial angle point.
Calculate the characteristic point of lip closure:
Left labial angle point, right labial angle point, lip pearl midpoint, lower interior labrale.
Calculate the characteristic point of mouth size:
Left cheekbone high point, right cheekbone high point, left labial angle point, right labial angle point.
Calculate the characteristic point of upper lip thin and thick:
Space between the eyebrows point, muffle angle point, lip valley point, lip pearl midpoint.
Calculate the characteristic point of lower lip thin and thick:
Space between the eyebrows point, muffle angle point, lower interior labrale, lower lip bottom point.
The characteristic point that each 90 ° of side face characteristic informations use is calculated specifically, being set forth below out, but it should know
That when actually calculating, can be calculated fully according to characteristic point disclosed in following, can also according to the actual situation into
Row fine tuning, but should at least use a part of corresponding characteristic point.
The characteristic point of calculating volume head trim:
Side metopion, side space between the eyebrows point.
Calculate the characteristic point of forehead plumpness:
The excessive point of volume, side volume end point under side metopion, side.
Calculate the characteristic point of space between the eyebrows height:
Side inferior crura of antihelix edge point, side lower eyelid outer rim bottom point, side space between the eyebrows point.
Calculate the characteristic point of eye convexity:
Branch hole socket of the eye vertex, side upper eyelid outer rim vertex.
Calculate the characteristic point of nasion height:
Side inferior crura of antihelix edge point, side lower eyelid outer rim bottom point, side nose mountain root point.
Calculate the characteristic point of nasion form:
Side inferior crura of antihelix edge point, side lower eyelid outer rim bottom point, side space between the eyebrows point, side nose mountain root point.
Calculate the characteristic point of contour of nasal bridge:
Side bridge of the nose starting point, side bridge of the nose high point, side nose starting point.
Calculate the characteristic point of nose height:
Side inferior crura of antihelix edge point, side lower eyelid outer rim bottom point, side corners of the mouth point, side nose vertex.
Calculate the high characteristic point of nose substrate:
Side inferior crura of antihelix edge point, side lower eyelid outer rim bottom point, side corners of the mouth point, side muffle angle point.
Calculate the characteristic point of nasal labial angle:
Side nose vertex, side muffle angle point, side upper lip are along vertex.
Calculate the convex flat characteristic point of upper lip:
Side inferior crura of antihelix edge point, side lower eyelid outer rim bottom point, side upper lip are along vertex, the excessive point of side chin, side muffle angle point.
Calculate the convex flat characteristic point of lower lip:
Side inferior crura of antihelix edge point, side lower eyelid outer rim bottom point, side upper lip under vertex, side lower lip the excessive point of inflection point, side chin,
Side muffle angle point.
Calculate the characteristic point of lower chin set-back:
Side inferior crura of antihelix edge point, side lower eyelid outer rim bottom point, side lower eyelid outer rim bottom point, side chin labial groove concave point, side chin are excessive
Point.
Calculate the characteristic point of lower chin head trim:
Side inferior crura of antihelix edge point, side lower eyelid outer rim bottom point, side lower eyelid outer rim bottom point, the excessive point of side chin.
Calculate the characteristic point of lower jaw edge clarity:
Pick up the ears hang down bottom point, side chin lower edge point, side neck jaw inflection point.
Calculate the characteristic point of jaw structure evaluation of estimate:
Side inferior crura of antihelix edge point, side lower eyelid outer rim bottom point, side corners of the mouth point.
The characteristic point that each 45 ° of side face characteristic informations use is calculated specifically, being set forth below out, but it should know
That when actually calculating, can be calculated fully according to characteristic point disclosed in following, can also according to the actual situation into
Row fine tuning, but should at least use a part of corresponding characteristic point.
Calculate the characteristic point of cheekbone height:
Tiltedly upper zygion, oblique cheekbone inflection point, oblique apple flesh high point.
Calculate the characteristic point of apple flesh plumpness:
Oblique cheekbone transition point, oblique apple flesh high point, tiltedly interior cheek starting point.
The characteristic point of cheek plumpness in calculating:
Cheek point, tiltedly interior cheek end point, oblique chin starting point in tiltedly.
In order to more accurately describe the makings of user, it can be and determine score feelings of the user in terms of different makings in advance
Condition, and then the description as described in terms of user's makings is provided in the makings score of different aspect according to user.
Specifically, step determines that the makings of user includes: according to face characteristic information
Face characteristic information is brought into preset makings model, to determine user in the reference value of each temperament type;
Makings model includes: the first makings model, the second makings model, third makings model, the 4th makings model, the 5th makings mould
Type;The corresponding temperament type of first makings model includes any one or more type below: affine, authoritative, mature, packet
Hold and wise;The corresponding temperament type of second makings model includes any one or more type below: face showing good fortune, luck for wealth,
Peach blossom, official career and longevity;The corresponding temperament type of third makings model includes any one or more type below: can
It leans on, is atmosphere, stingy, agile and simple-minded;The corresponding temperament type of 4th makings model include it is below any one or more
Type: bitter phase, native phase, is entangled with and nurse a grievance the appearance or manner suggestive of abject poverty;The corresponding temperament type of 5th makings model include it is below any one
Or multiple types: cold and detached, weak, young, unkind and slow-witted;
According to reference value of the user in terms of each temperament type, the makings description of user is determined.
More specifically, affine be calculated according to any one or more characteristic informations below: eye is long, lip is raised
Angle, mouth size, convex flat, the lower jaw edge clarity of upper lip, apple flesh plumpness;
Authority is calculated according to any one or more characteristic informations below: symmetry, frontal temporal part, eyebrow are raised
Angle, bizygomatic, cheekbone height, space between the eyebrows height, contour of nasal bridge, jaw be wide, lower chin head trim;
Maturation is calculated according to any one or more characteristic informations below: length-width ratio, lower presiding judge's degree, symmetrical
Degree, eye size, eye spacing, nose substrate is high, jaw is wide, lower chin set-back;
Containing is calculated according to any one or more characteristic informations below: upper presiding judge's degree, frontal temporal part, temporo
Wide, glabella is away from, eye spacing, wing of nose width, mouth size, jaw are wide, lower chin head trim;
Wise be calculated according to any one or more characteristic informations below: upper presiding judge's degree, frontal temporal part, forehead
Plumpness, temporo are wide, eye is long, eye spacing, space between the eyebrows height, lower chin set-back;
Face showing good fortune is calculated according to any one or more characteristic informations below: upper presiding judge's degree, temporo is wide, eye is long,
Temporo is wide, eye is long, wing of nose width, lip raise angle, upper lip thin and thick, jaw are wide, lower chin head trim, outer cheek plumpness, apple flesh plumpness;
Luck for wealth is calculated according to any one or more characteristic informations below: frontal temporal part, nasion height, nose
Highly, wing of nose width, jaw be wide, lower chin set-back;
Peach blossom is calculated according to any one or more characteristic informations below: symmetry, frontal temporal part, forehead are full
Full scale, glabella away from, eye size, eye raise angle, lip raises angle, upper lip thin and thick, lower lip thin and thick, lower chin length, apple flesh plumpness;
Official career is calculated according to any one or more characteristic informations below: upper presiding judge's degree, symmetry, volume temporo
Portion, space between the eyebrows height, nose height, nose substrate height, wing of nose width, jaw be wide, lower chin set-back, lower chin head trim;
Longevity is calculated according to any one or more characteristic informations below: lower presiding judge's degree, eyebrow tail length, the nasion
Highly, wing of nose width, length, lower chin head trim in people;
It is reliably to be calculated according to any one or more characteristic informations below: frontal temporal part, nose substrate height, the wing of nose
Width, jaw be wide, lower chin set-back, lower chin head trim;
Atmosphere is calculated according to any one or more characteristic informations below: frontal temporal part, temporo be wide, glabella away from,
Eye size, eye spacing, wing of nose width, mouth size, jaw are wide;
Stingy be calculated according to any one or more characteristic informations below: frontal temporal part, temporo be wide, glabella away from,
Eye size, eye spacing, wing of nose width, mouth size, jaw are wide;
Agile be calculated according to any one or more characteristic informations below: length-width ratio, volume head trim, nose base
Bottom height, wing of nose width, upper lip thin and thick, lower lip thin and thick, jaw be wide, lower chin set-back;
Simple-minded be calculated according to any one or more characteristic informations below: length-width ratio, volume head trim, nose base
Bottom height, wing of nose width, upper lip thin and thick, lower lip thin and thick, jaw be wide, lower chin set-back;
Hardship is calculated according to any one or more characteristic informations below: upper presiding judge's degree, temporo be wide, glabella
Away from, eye, long, wing of nose width, lip raise wide angle, upper lip thin and thick, jaw, lower chin head trim, outer cheek plumpness, apple flesh plumpness;
The appearance or manner suggestive of abject poverty is calculated according to any one or more characteristic informations below: frontal temporal part, nasion height, nose
Highly, wing of nose width, jaw be wide, lower chin set-back;
Soil is calculated according to any one or more characteristic informations below: before upper presiding judge's degree, frontal temporal part, volume
Inclination, looks spacing, eye size, space between the eyebrows height, nasion form, contour of nasal bridge, nose height, nose substrate height, wing of nose width, under
Lip is convex flat;
Be entangled with is calculated according to any one or more characteristic informations below: face distribution, frontal temporal part, glabella
Away from, eye spacing, nasion form;
Nurse a grievance is calculated according to any one or more characteristic informations below: nose substrate is high, upper lip it is convex it is flat, under
Chin head trim;
Indifferently to be calculated according to any one or more characteristic informations below: eye is long, lip raises angle, mouth size,
Convex flat, the lower jaw edge clarity of upper lip, apple flesh plumpness;
Weak be calculated according to any one or more characteristic informations below: frontal temporal part, eyebrow raise angle, bizygomatic,
Cheekbone height, space between the eyebrows height, contour of nasal bridge, jaw be wide, lower chin head trim;
Youth is calculated according to any one or more characteristic informations below: length-width ratio, lower presiding judge's degree, symmetrical
Degree, eye size, eye spacing, nose substrate height, lower chin set-back;
Unkind be calculated according to any one or more characteristic informations below: upper presiding judge's degree, frontal temporal part, temporo
Wide, glabella away from, eye spacing, wing of nose width, in people the convex flat, lower lip of length, mouth size, upper lip it is convex it is flat, jaw is wide;
Slow-witted be calculated according to any one or more characteristic informations below: upper presiding judge's degree, frontal temporal part, forehead
Plumpness, temporo are wide, eye is long, eye spacing, space between the eyebrows height, lower chin set-back.
Wherein, the reference value of temperament type has mainly reacted user in the scoring of different temperament types, later, according to commenting
Divide the makings description determined to user.When practical operation, a tables of data can be pre-established, is recorded in the tables of data
The corresponding relationship of reference value and the makings description of temperament type, in turn, after the reference value of temperament type has been determined, Ke Yizhi
It connects and transfers corresponding makings description from tables of data as the comment to user.As shown in table 4 below, some gas of crucial point
The relationship of the scoring of matter type and the description of corresponding makings, it is corresponding, one can be established for each temperament type and is corresponded to
Makings description corresponding relationship, be also possible to for all temperament types to be all integrated together foundation and the description of corresponding makings
Relationship, that is, it is X that the scoring of A temperament type has been reacted in table, and the scoring of B temperament type is Y, the scoring of C temperament type
When for Z, what corresponding makings description is.
Table 4
As shown in table 5 below, the corresponding relationship of face characteristic information and temperament type is shown;It has been reflected often in other words
The calculated relationship of the reference value of a temperament type and corresponding face characteristic information, as shown in table 4, the reference value of affinity is
, nose height long according to eye, nostril, lip raise angle, mouth size, lower jaw edge clarity, outer cheek plumpness and apple flesh plumpness
It is calculated.The numbers such as 1,2 and 3 therein show weight, that is, apple flesh plumpness is 3, and the weight that lip raises angle is
2, remaining weight is 1.
Table 5
Carry out the calculating process of the positive face characteristic information of declaratives with specific several examples below:
Example 1: the mode of eye size is calculated:
As shown in figure 5, having marked following characteristic point in figure: MEB (metopion), LN1 (left frontal eminence horizontal extension point),
RN1 (right frontal eminence horizontal extension point), LQ1 (left cheekbone high point), RQ1 (right cheekbone high point), LH1 (left jaw angle point), RH1 (right jaw
Angle point), MKB (chin bottom point), LY1 (left eye endocanthion point), RY1 (right eye endocanthion point), LY3 (upper left margo palpebrae vertex), RY3 (upper right
Margo palpebrae vertex), LY5 (left eye outer canthus point), RY5 (right eye outer canthus point), LY7 (lower-left margo palpebrae bottom point), (the bottom right margo palpebrae bottom RY7
Point).
Calculation formula:
Average palpebral fissure height:
YH=
[(LY3-LY7)+(RY3-RY7)]/2=[((971,698)-(979,760))+(and (602,692)-(604,
755))]/2≈125.21;
Average eye is wide:
YW=[(LY5-LY1)+(RY1-RY5)]/2=[((1056,736)-(900,747))+(and (675,748)-(513,
727))]/2≈320.01;
Face area:
MS=S (MEB, LN1, LQ1, LH1, MKB, RH1, RQ1, RN1)=[(780,283), (1117,424), (1198,
788), (1117,1142), (799,1393), (448,1143), (362,789), (438,420)]=560250.47;
Eye area: YS=2 × YH × YW ≈ 20034.28;
Eye size characteristic value K=100 × YS/MS ≈ 3.58;
It should be noted that calculated characteristic value (such as eye size characteristic value), eye area, numerical value all can serve as
The characteristic information of eye size depending on which kind of specifically used numerical value can be according to actual conditions as final characteristic information, but passes through
Cross the actual use and verifying of inventor, it is believed that characteristic value is the most suitable as corresponding characteristic information.It is i.e. that eye size is special
Value indicative is the most suitable as the characteristic information of eye size, similar, hereinafter in exemplifications set out, it is also preferred that characteristic value is made
For corresponding characteristic information, but other numerical value also can be used as characteristic information.
Example 2: the mode of Zhongting length is calculated:
As shown in fig. 6, having marked following characteristic point in figure:
MEB (metopion), MYT (space between the eyebrows point), MBC (muffle angle point), MKB (chin bottom point);
Calculation formula:
The long LL=MEB-MKB=of face (780,283)-(799,1393) ≈ 1110.85;
Zhongting length ZL=MYT-MBC=(788,624)-(790,1034) ≈ 409.7;
Zhongting length characteristic value K=ZL/LL ≈ 0.37;
Example 3: the mode of bizygomatic is calculated:
As shown in fig. 7, having marked following characteristic point in figure:
RN3 (right temples concave point), LN3 (left temples concave point), RQ1 (right cheekbone high point), LQ1 (left cheekbone high point), RW1
(right cheekbone lower edge point), LW1 (left cheekbone lower edge point);
The wide NW=RN3-LN3=of temporo (360,620)-(1197,644) ≈ 838.16;
The wide LW=RQ1-LQ1=of face (362,789)-(1198,788) ≈ 835.3;
The outer wide WW=RW1-LW1=of cheek (397,972)-(1162,978) ≈ 764.62;
Bizygomatic characteristic value K=(NW+WW)/2/LW ≈ 1;
Carry out the calculating process of 90 ° of side face characteristic informations of declaratives with specific several examples below:
Example 4: the high mode of nose substrate is calculated:
As shown in figure 8, having marked following characteristic point in figure:
CE1 (side inferior crura of antihelix edge point), CY3 (side lower eyelid outer rim vertex), CBJ (side nose basal point), CC5 (the side corners of the mouth
Point);
EE line (CE1 point and CY3 point line) length: EE=CE1-CY3 ≈ 490.43;
EL line (CY3 point and CC5 point line);
BJ line (CBJ point to EL line) length: BJ ≈ 62.13;
The high characteristic value of nose substrate: K=BJ/EE ≈ 7.89;
Example 5: the mode of nasal labial angle is calculated:
As shown in figure 9, having marked following characteristic point in figure:
CZ1 (side nose pillar midpoint), CZ2 (side muffle angle point), CRZ (midpoint in people from side);
Nasal labial angle: the line of point CZ1 and point CZ2, with included angle A on the right side of the line of point CZ2 and point CRZ;
Nasal labial angle characteristic value: 104.14 ° of A ≈.
Illustrate the calculating process of 45 ° of side face characteristic informations of groups of people's face with specific several examples below:
Example 6: the mode of cheekbone height is calculated:
As shown in Figure 10, following characteristic point has been marked in figure:
XQ1 (tiltedly upper zygion), XQ2 (oblique cheekbone inflection point), XQ4 (oblique apple flesh high point);
Cheekbone height: the line of point XQ1 and point XQ2, with included angle A on the left of the line of point XQ2 and point XQ4;
Cheekbone altitude feature value: 158.74 ° of A ≈;
Example 7: the mode of apple flesh plumpness is calculated:
As shown in figure 11, following characteristic point has been marked in figure:
XQ3 (oblique cheekbone transition point), XQ4 (oblique apple flesh high point), XN1 (cheek starting point in tiltedly);
Apple flesh plumpness: the line of point XQ3 and point XQ4, with included angle A on the left of the line of point XQ4 and point XN1;
Apple flesh plumpness characteristic value: 175.49 ° of A ≈.
Training characteristics identification model process is briefly described below:
Step 201, multiple training samples are obtained;Characteristic point (the artificial mark manually marked out is recorded on training sample
The mode of characteristic point can be with reference to the position in above-mentioned table 1-3);
Step 202, a random training sample is input to not training to complete in model, to determine training result;
Step 203, according to the characteristic point in the characteristic point and training result manually marked out on training sample, training is calculated
As a result corresponding loss function;
Step 204, judge whether loss function is less than preset threshold value;If so, 205 are thened follow the steps, if it is not, then executing
Step 206;
Step 205, current not training model is completed as feature identification model to export;
Step 206, model is completed to not training according to loss function to be trained, and re-execute the steps 202.
It corresponds to the above method, present invention also provides a kind of user's makings determining devices, comprising:
Module is obtained, for obtaining the two-dimentional face image of user;
Input module, for two-dimentional face image to be input to feature identification model, to determine facial feature points;
Computing module, for calculating face characteristic information according to facial feature points;
Determining module, for determining the makings of user according to face characteristic information.
It corresponds to the above method, present invention also provides a kind of non-volatile program generations that can be performed with processor
The computer-readable medium of code, which is characterized in that said program code makes the processor execute a kind of user's makings determination side
Method.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.
And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic or disk.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. a kind of user's makings determines method characterized by comprising
Obtain the two-dimentional face image of user;
Two-dimentional face image is input to feature identification model, to determine facial feature points;
Face characteristic information is calculated according to facial feature points;
The makings of user is determined according to face characteristic information.
2. the method according to claim 1, wherein the two-dimentional facial photo that step obtains user includes:
Obtain three-dimensional face image;Three-dimensional face image is to be scanned to obtain using face of the three-dimensional scanner to user
's;
The mapping of specified angle is carried out, to three-dimensional face image to determine two-dimentional face image;It is described two dimension face image include
Any one or more image below: face image, 90 ° of side face images and 45 ° of side face images.
3. according to the method described in claim 2, it is characterized in that, facial feature points include any one or more spies below
Syndrome group: Zheng Lian feature group, 90 ° of side face feature groups and 45 ° of side face feature groups;
Zheng Lian feature group includes any one or more characteristic point groups below: outer profile point group, positive eye point group group,
Nose point group and mouth point group;
90 ° of side face feature groups include any one or more characteristic point groups below:
Side profile point group, branch hole Bu Dian group and point group of ear;
45 ° of side face feature groups include any one or more characteristic point groups below:
Oblique profile point group;
Outer profile point group includes following any one or more characteristic points:
Metopion, left frontal eminence horizontal extension point, right frontal eminence horizontal extension point, left temples salient point, right temples salient point, left temples are recessed
Point, right temples concave point, left cheekbone high point, right cheekbone high point, left cheekbone lower edge point, right cheekbone lower edge point, left jaw angle point, right jaw angle
Point and chin bottom point;
Positive eye point group group includes following any one or more characteristic points:
Upper left brows starting point, upper right brows starting point, left eyebrow tail point, right eyebrow tail point, upper left eye socket vertex, upper right eye socket vertex, a left side
Intraocular corner of the eyes point, right eye endocanthion point, upper left margo palpebrae vertex, upper right margo palpebrae vertex, left eye outer canthus point, right eye outer canthus point, lower-left margo palpebrae
Bottom point and bottom right margo palpebrae bottom point;
Nose point group includes following any one or more characteristic points:
Space between the eyebrows point, left nasal base breadth point, right nasal base breadth point, left nose hole vertex, right nostril vertex and muffle angle point;
Mouth point group includes following any one or more characteristic points:
Lip valley point, left labial angle point, right labial angle point, lower lip bottom point, lip pearl midpoint and lower interior labrale;
Side profile point group includes following any one or more characteristic points:
The excessive point of volume under side metopion, side, side volume end point, side space between the eyebrows point, side nose mountain root point, side bridge of the nose starting point, side bridge of the nose high point,
Side nose starting point, side nose vertex, side nostril vertex, side nose pillar midpoint, side muffle angle point, midpoint, side upper lip edge in people from side
Inflection point, side chin labial groove concave point, the excessive point of side chin, side chin lower edge point, side neck jaw inflection point under vertex, side lower lip;
Branch hole Bu Dian group includes following any one or more characteristic points:
Branch hole socket of the eye vertex, side upper eyelid outer rim vertex, side lower eyelid outer rim bottom point, side nose basal point and side corners of the mouth point;
Point group of ear includes following any one or more characteristic points:
Side inferior crura of antihelix edge point and bottom point of hanging down of picking up the ears;
Oblique profile point group includes following any one or more characteristic points:
Tiltedly upper zygion, oblique cheekbone inflection point, oblique cheekbone transition point, oblique apple flesh high point, tiltedly in cheek starting point, tiltedly in cheek point, tiltedly
Interior cheek end point and oblique chin starting point.
4. according to the method described in claim 3, it is characterized in that, face characteristic information includes below any one or more
Characteristic information:
Positive face characteristic information, 90 ° of side face characteristic informations and 45 ° of side face characteristic informations;
Positive face characteristic information includes below any one or more:
Face's length-width ratio, upper presiding judge's degree, face amount is wide, temporo is wide, bizygomatic, outer cheek plumpness, jaw are wide, lower presiding judge's degree, lower chin length,
Symmetry, face Evaluation on distribution value, eyebrow tail length, eyebrow raise angle, glabella away from, looks spacing, eye is long, palpebral fissure size, eye size, eye are raised
Angle, eye spacing, Zhongting length, wing of nose width, the exposed degree in nostril, to raise angle, lip closure, mouth big for length, lip separated time, lip in people
Small, upper lip thin and thick, lower lip thin and thick, jaw structure evaluation of estimate;
90 ° of side face characteristic informations include below any one or more:
Volume head trim, forehead plumpness, space between the eyebrows height, eye convexity, nasion height, nasion form, contour of nasal bridge, nose height,
Convex flat, convex flat, the lower chin set-back of lower lip of nose substrate height, nasal labial angle, upper lip, lower chin head trim, lower jaw edge clarity, jaw structure are commented
Value;
45 ° of side face characteristic informations include below any one or more:
Cheekbone height, apple flesh plumpness, interior cheek plumpness.
5. according to the method described in claim 4, it is characterized in that, step calculates face characteristic information packet according to facial feature points
Include following any one or more steps:
According to the characteristic point in Zheng Lian feature group, positive face characteristic information is calculated;
According to the characteristic point in 90 ° of side face feature groups, 90 ° of side face characteristic informations are calculated;
According to the characteristic point in 45 ° of side face feature groups, 45 ° of side face characteristic informations are calculated.
6. according to the method described in claim 3, it is characterized in that, two-dimentional face image is input to feature identification mould by step
Type, to determine that facial feature points include following any one or more steps:
Two-dimentional positive face face image is input to feature identification model, to determine that Zheng Lian feature group, Zheng Lian feature group include
Characteristic point in any one or more groups below: outer profile point group, positive eye point group group, nose point group and mouth point
Group;
90 ° of side face images are input to feature identification model, with determining 90 ° of side face feature groups, 90 ° of side face feature groups include
Characteristic point in any one or more groups below: side profile point group, branch hole Bu Dian group, point group of ear;
The third mode: being input to feature identification model for 45 ° of side face images, to determine 45 ° of side face feature groups, 45 ° of side faces
Feature group includes oblique profile point group.
7. the method according to claim 1, wherein step determines the makings packet of user according to face characteristic information
It includes:
Face characteristic information is brought into preset makings model, to determine user in the reference value of each temperament type;Makings
Model includes: the first makings model, the second makings model, third makings model, the 4th makings model, the 5th makings model;The
The corresponding temperament type of one makings model includes any one or more type below: it is affine, authoritative, mature, contain and
It is wise;The corresponding temperament type of second makings model includes any one or more type below: face showing good fortune, luck for wealth, peach blossom,
Official career and longevity;The corresponding temperament type of third makings model includes any one or more type below: reliable, big
It is gas, stingy, agile and simple-minded;The corresponding temperament type of 4th makings model includes any one or more type below:
Bitter phase, native phase, is entangled with and nurse a grievance the appearance or manner suggestive of abject poverty;The corresponding temperament type of 5th makings model include it is below any one or it is more
Seed type: cold and detached, weak, young, unkind and slow-witted;
According to reference value of the user in terms of each temperament type, the makings description of user is determined.
8. the method according to the description of claim 7 is characterized in that
Affine be calculated according to any one or more characteristic informations below: eye is long, lip raises angle, mouth size, upper lip
Convex flat, lower jaw edge clarity, apple flesh plumpness;
Authority is calculated according to any one or more characteristic informations below: symmetry, frontal temporal part, eyebrow raise angle, cheekbone
Width, cheekbone height, space between the eyebrows height, contour of nasal bridge, jaw be wide, lower chin head trim;
Maturation is calculated according to any one or more characteristic informations below: length-width ratio, lower presiding judge's degree, symmetry,
Eye size, eye spacing, nose substrate is high, jaw is wide, lower chin set-back;
Containing is calculated according to any one or more characteristic informations below: upper presiding judge's degree, frontal temporal part, temporo be wide, eyebrow
Spacing, eye spacing, wing of nose width, mouth size, jaw be wide, lower chin head trim;
Wise be calculated according to any one or more characteristic informations below: upper presiding judge's degree, frontal temporal part, forehead are full
Degree, temporo is wide, eye is long, eye spacing, space between the eyebrows height, lower chin set-back;
Face showing good fortune is calculated according to any one or more characteristic informations below: upper presiding judge's degree, temporo is wide, eye is long, temporo is wide,
The long, wing of nose width of eye, lip raise wide angle, upper lip thin and thick, jaw, lower chin head trim, outer cheek plumpness, apple flesh plumpness;
Luck for wealth is calculated according to any one or more characteristic informations below: frontal temporal part, nasion height, nose are high
Degree, wing of nose width, jaw be wide, lower chin set-back;
Peach blossom is calculated according to any one or more characteristic informations below: symmetry, frontal temporal part, forehead are full
Degree, glabella away from, eye size, eye raise angle, lip raises angle, upper lip thin and thick, lower lip thin and thick, lower chin length, apple flesh plumpness;
Official career is calculated according to any one or more characteristic informations below: upper presiding judge's degree, symmetry, frontal temporal part,
Space between the eyebrows height, nose height, nose substrate height, wing of nose width, jaw be wide, lower chin set-back, lower chin head trim;
Longevity is calculated according to any one or more characteristic informations below: lower presiding judge's degree, eyebrow tail length, nasal root salient
Degree, wing of nose width, length, lower chin head trim in people;
It is reliably to be calculated according to any one or more characteristic informations below: frontal temporal part, nose substrate height, nasal base breadth
It spends, jaw is wide, lower chin set-back, lower chin head trim;
Atmosphere is calculated according to any one or more characteristic informations below: frontal temporal part, temporo be wide, glabella is big away from, eye
Small, eye spacing, wing of nose width, mouth size, jaw are wide;
Stingy be calculated according to any one or more characteristic informations below: frontal temporal part, temporo be wide, glabella is big away from, eye
Small, eye spacing, wing of nose width, mouth size, jaw are wide;
Agile be calculated according to any one or more characteristic informations below: length-width ratio, volume head trim, nose substrate
Height, wing of nose width, upper lip thin and thick, lower lip thin and thick, jaw be wide, lower chin set-back;
Simple-minded be calculated according to any one or more characteristic informations below: length-width ratio, volume head trim, nose substrate
Height, wing of nose width, upper lip thin and thick, lower lip thin and thick, jaw be wide, lower chin set-back;
Hardship is calculated according to any one or more characteristic informations below: upper presiding judge's degree, temporo be wide, glabella is away from, eye
Length, wing of nose width, lip raise wide angle, upper lip thin and thick, jaw, lower chin head trim, outer cheek plumpness, apple flesh plumpness;
The appearance or manner suggestive of abject poverty is calculated according to any one or more characteristic informations below: frontal temporal part, nasion height, nose are high
Degree, wing of nose width, jaw be wide, lower chin set-back;
Soil is calculated according to any one or more characteristic informations below: upper presiding judge's degree, frontal temporal part, volume lean forward
Degree, looks spacing, eye size, space between the eyebrows height, nasion form, contour of nasal bridge, nose height, nose substrate height, wing of nose width, lower lip
It is convex flat;
Be entangled with is calculated according to any one or more characteristic informations below: face distribution, frontal temporal part, glabella away from,
Eye spacing, nasion form;
Nurse a grievance is calculated according to any one or more characteristic informations below: before nose substrate height, convex flat, the lower chin of upper lip
Inclination;
Be indifferently to be calculated according to any one or more characteristic informations below: eye is long, lip raises angle, mouth size, upper lip
Convex flat, lower jaw edge clarity, apple flesh plumpness;
Weak be calculated according to any one or more characteristic informations below: frontal temporal part, eyebrow raise angle, bizygomatic, cheekbone
Highly, space between the eyebrows height, contour of nasal bridge, jaw be wide, lower chin head trim;
Youth is calculated according to any one or more characteristic informations below: length-width ratio, lower presiding judge's degree, symmetry,
Eye size, eye spacing, nose substrate height, lower chin set-back;
Unkind be calculated according to any one or more characteristic informations below: upper presiding judge's degree, frontal temporal part, temporo be wide, eyebrow
Spacing, eye spacing, wing of nose width, in people the convex flat, lower lip of length, mouth size, upper lip it is convex it is flat, jaw is wide;
Slow-witted be calculated according to any one or more characteristic informations below: upper presiding judge's degree, frontal temporal part, forehead are full
Degree, temporo is wide, eye is long, eye spacing, space between the eyebrows height, lower chin set-back.
9. a kind of user's makings determining device characterized by comprising
Module is obtained, for obtaining the two-dimentional face image of user;
Input module, for two-dimentional face image to be input to feature identification model, to determine facial feature points;
Computing module, for calculating face characteristic information according to facial feature points;
Determining module, for determining the makings of user according to face characteristic information.
10. a kind of computer-readable medium for the non-volatile program code that can be performed with processor, which is characterized in that described
Program code makes the processor execute described any the method for claim 1-8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811045713.4A CN109255328A (en) | 2018-09-07 | 2018-09-07 | User's makings determines method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811045713.4A CN109255328A (en) | 2018-09-07 | 2018-09-07 | User's makings determines method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109255328A true CN109255328A (en) | 2019-01-22 |
Family
ID=65047043
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811045713.4A Pending CN109255328A (en) | 2018-09-07 | 2018-09-07 | User's makings determines method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109255328A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020183224A1 (en) * | 2019-03-12 | 2020-09-17 | Sheikh Roshandel Hamidreza | A system and method for relative interpretation and assessment of innate temperament of individuals using face parts' ratios |
CN112052734A (en) * | 2020-07-31 | 2020-12-08 | 北京奇艺世纪科技有限公司 | Role characteristic identification network generation method, role characteristic identification device, role characteristic identification equipment and role characteristic identification medium |
CN112836545A (en) * | 2019-11-22 | 2021-05-25 | 北京新氧科技有限公司 | 3D face information processing method and device and terminal |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101604377A (en) * | 2009-07-10 | 2009-12-16 | 华南理工大学 | A kind of facial beauty classification method that adopts computing machine to carry out woman image |
CN106991371A (en) * | 2017-03-01 | 2017-07-28 | 浙江大学 | A kind of face type collection and analysis system based on traditional Chinese medical science five-element people's classification of TCM constitution |
CN107315987A (en) * | 2016-04-27 | 2017-11-03 | 伽蓝(集团)股份有限公司 | Assess facial apparent age, the method and its application of facial aging degree |
CN107729882A (en) * | 2017-11-19 | 2018-02-23 | 济源维恩科技开发有限公司 | Emotion identification decision method based on image recognition |
CN107798318A (en) * | 2017-12-05 | 2018-03-13 | 四川文理学院 | The method and its device of a kind of happy micro- expression of robot identification face |
CN108304811A (en) * | 2018-02-07 | 2018-07-20 | 人加智能机器人技术(北京)有限公司 | Three-dimensional face face value methods of marking, apparatus and system based on RGBD information |
-
2018
- 2018-09-07 CN CN201811045713.4A patent/CN109255328A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101604377A (en) * | 2009-07-10 | 2009-12-16 | 华南理工大学 | A kind of facial beauty classification method that adopts computing machine to carry out woman image |
CN107315987A (en) * | 2016-04-27 | 2017-11-03 | 伽蓝(集团)股份有限公司 | Assess facial apparent age, the method and its application of facial aging degree |
CN106991371A (en) * | 2017-03-01 | 2017-07-28 | 浙江大学 | A kind of face type collection and analysis system based on traditional Chinese medical science five-element people's classification of TCM constitution |
CN107729882A (en) * | 2017-11-19 | 2018-02-23 | 济源维恩科技开发有限公司 | Emotion identification decision method based on image recognition |
CN107798318A (en) * | 2017-12-05 | 2018-03-13 | 四川文理学院 | The method and its device of a kind of happy micro- expression of robot identification face |
CN108304811A (en) * | 2018-02-07 | 2018-07-20 | 人加智能机器人技术(北京)有限公司 | Three-dimensional face face value methods of marking, apparatus and system based on RGBD information |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020183224A1 (en) * | 2019-03-12 | 2020-09-17 | Sheikh Roshandel Hamidreza | A system and method for relative interpretation and assessment of innate temperament of individuals using face parts' ratios |
CN112836545A (en) * | 2019-11-22 | 2021-05-25 | 北京新氧科技有限公司 | 3D face information processing method and device and terminal |
CN112052734A (en) * | 2020-07-31 | 2020-12-08 | 北京奇艺世纪科技有限公司 | Role characteristic identification network generation method, role characteristic identification device, role characteristic identification equipment and role characteristic identification medium |
CN112052734B (en) * | 2020-07-31 | 2024-05-17 | 北京奇艺世纪科技有限公司 | Character characteristic recognition network generation method, character characteristic recognition network generation program, character characteristic recognition apparatus, device and medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP7399210B2 (en) | Method for processing ocular images of the eyes of a wearer of a head-mounted display system | |
KR102342982B1 (en) | Methods and related augmented reality methods for concealing objects in images or videos | |
CN103443827B (en) | For providing the method and system of face adjustment image | |
CN106030661B (en) | The independent 3D scene texture background in the visual field | |
US10201273B2 (en) | Method for determining ocular measurements using a consumer sensor | |
JP6074494B2 (en) | Shape recognition device, shape recognition program, and shape recognition method | |
CN108921148A (en) | Determine the method and device of positive face tilt angle | |
CN109255328A (en) | User's makings determines method and device | |
CN105938627A (en) | Processing method and system for virtual plastic processing on face | |
JP4692526B2 (en) | Gaze direction estimation apparatus, gaze direction estimation method, and program for causing computer to execute gaze direction estimation method | |
CN106909875A (en) | Face shape of face sorting technique and system | |
CN109145866A (en) | Determine the method and device of side face tilt angle | |
CN109255327A (en) | Acquisition methods, face's plastic operation evaluation method and the device of face characteristic information | |
CN110188728A (en) | A kind of method and system of head pose estimation | |
US20150243015A1 (en) | Method and device for preparing a spectacle frame | |
CN109145865A (en) | Face standard level calculating method and device | |
CN109785396B (en) | Writing posture monitoring method, system and device based on binocular camera | |
CN105874378A (en) | Method for determining a geometric definition of a customized optical device | |
JP2022538669A (en) | Improved eye tracking latency | |
CN106650654B (en) | A kind of three-dimensional hair line extracting method based on human body head colour point clouds model | |
CN104408462A (en) | Quick positioning method of facial feature points | |
CN112220444A (en) | Pupil distance measuring method and device based on depth camera | |
CN109241911A (en) | Human face similarity degree calculation method and device | |
CN112733579A (en) | Object reconstruction method, device, equipment and storage medium | |
CN106570747A (en) | Glasses online adaption method and system combining hand gesture recognition |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190122 |
|
RJ01 | Rejection of invention patent application after publication |