CN108985657A - Evaluation method, device and the server of personage's clothing collocation - Google Patents
Evaluation method, device and the server of personage's clothing collocation Download PDFInfo
- Publication number
- CN108985657A CN108985657A CN201810887103.2A CN201810887103A CN108985657A CN 108985657 A CN108985657 A CN 108985657A CN 201810887103 A CN201810887103 A CN 201810887103A CN 108985657 A CN108985657 A CN 108985657A
- Authority
- CN
- China
- Prior art keywords
- clothing
- sample
- target object
- body type
- collocation
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Professional, Industrial, Or Sporting Protective Garments (AREA)
Abstract
The embodiment of the present invention provides evaluation method, device and the server of a kind of personage's clothing collocation, is applied to augmented reality, this method comprises: first obtaining the body information and present cloth information of target object；And the body type of target object is determined according to body information；The body type of target object and present cloth information input are obtained into the evaluation result of target object clothing collocation into clothing collocation evaluation model again.Evaluation method, device and the server of personage's clothing collocation provided in an embodiment of the present invention, improve the accuracy of clothing collocation evaluation.
Description
Technical field
The present invention relates to evaluation method, device and clothes that field of computer technology more particularly to a kind of personage's clothing are arranged in pairs or groups
Business device.
Background technique
In the prior art, certain an object is arranged in pairs or groups to clothing, or when evaluating the clothing matched, usually
Be it is aesthetic according to the subjectivity of the object, arrange in pairs or groups to clothing, or the clothing matched is evaluated, but due to each
The aesthetic difference of subjectivity of object, when so that evaluating for the same clothing matched, obtained evaluation result is different, from
And cause the accuracy of clothing collocation evaluation not high.
Summary of the invention
The embodiment of the present invention provides evaluation method, device and the server of a kind of personage's clothing collocation, is taken with improving clothing
Accuracy with evaluation.
In a first aspect, the embodiment of the present invention provides a kind of evaluation method of personage's clothing collocation, the method is for enhancing
Reality, which comprises
Obtain the body information and present cloth information of target object；
The body type of the target object is determined according to the body information；
By the body type of the target object and the present cloth information input into clothing collocation evaluation model, obtain
The evaluation result arranged in pairs or groups to the target object clothing.
In one possible implementation, the clothing collocation evaluation model is trained by the following method:
According to the body type in training sample each in multiple training samples, the corresponding fisrt feature of each sample is constructed
Vector；
According to the corresponding first eigenvector of each sample and the clothing information marked in each sample, the clothing is determined
Object collocation evaluation model.
In one possible implementation, described according to the corresponding first eigenvector of each sample and in each sample
The clothing information of middle label determines the clothing collocation evaluation model, comprising:
Using deep learning method to the clothing each sample corresponding first eigenvector and marked in each sample
Information is trained, and obtains the clothing collocation evaluation model.
In one possible implementation, the body class that the target object is determined according to the body information
Type, comprising:
Obtain body type detection model；
By the body information input of the target object into the body type detection model, the target object is obtained
Body type.
In one possible implementation, the acquisition body type detection model, comprising:
According to the body information in training sample each in multiple training samples, the corresponding second feature of each sample is constructed
Vector；
According to the body type that the corresponding second feature vector sum of each sample marks in each sample, the shape is determined
Body type detection model.
In one possible implementation, it is described according to the corresponding second feature vector sum of each sample in each sample
The body type of middle label determines the body type detection model, comprising:
The body that the corresponding second feature vector sum of each sample is marked in each sample using deep learning method
Type is trained, and obtains the body type detection model.
In one possible implementation, described by the body type of the target object and the present cloth information
It is input in clothing collocation evaluation model, after obtaining the evaluation result of target object clothing collocation, further includes:
According to the corresponding relationship between the body type of the target object and body type and clothing collocation strategy, really
Fixed corresponding clothing collocation strategy；
The body type for sending evaluation result and the target object that the target object clothing is arranged in pairs or groups to terminal is corresponding
Clothing arrange in pairs or groups strategy.
Second aspect, the embodiment of the present invention also provide a kind of evaluating apparatus of personage's clothing collocation, and described device is for increasing
Strong reality, described device include:
Acquiring unit, for obtaining the body information and present cloth information of target object；
Processing unit, for determining the body type of the target object according to the body information；
The processing unit is also used to the body type of the target object and the present cloth information input to clothing
Object is arranged in pairs or groups in evaluation model, and the evaluation result of the target object clothing collocation is obtained.
In one possible implementation, described device further includes construction unit, and the clothing collocation evaluation model is
It trains by the following method:
The construction unit, for according to the body type in training sample each in multiple training samples, building to be each
The corresponding first eigenvector of sample；
The processing unit is also used to mark according to the corresponding first eigenvector of each sample and in each sample
Clothing information determines the clothing collocation evaluation model.
In one possible implementation, the processing unit is specifically used for using deep learning method to each sample
This corresponding first eigenvector and the clothing information marked in each sample are trained, and obtain the clothing collocation evaluation
Model.
In one possible implementation, the processing unit is specifically used for obtaining body type detection model；And it will
The body information input of the target object obtains the body class of the target object into the body type detection model
Type.
In one possible implementation, construction unit is also used to according to training sample each in multiple training samples
In body information, construct the corresponding second feature vector of each sample；
The processing unit is also used to be marked in each sample according to the corresponding second feature vector sum of each sample
Body type determines the body type detection model.
In one possible implementation, the processing unit is specifically used for using deep learning method to each sample
The body type that this corresponding second feature vector sum marks in each sample is trained, and obtains the body type detection
Model.
In one possible implementation, described device further includes transmission unit；
The processing unit is also used to plan of arranging in pairs or groups according to the body type and body type and clothing of the target object
Corresponding relationship between slightly determines corresponding clothing collocation strategy；
The transmission unit, for sending the evaluation result and the target that the target object clothing is arranged in pairs or groups to terminal
The corresponding clothing collocation strategy of the body type of object.
The third aspect, the embodiment of the present invention also provide a kind of server, including processor and memory, wherein described to deposit
Reservoir is for storing program instruction；
The processor is used to read the program instruction in the memory, and according to the program instruction in the memory
Execute the evaluation method of the collocation of personage's clothing shown in above-mentioned any one of first aspect.
Fourth aspect, the embodiment of the present invention also provide a kind of computer readable storage medium, computer readable storage medium
On be stored with computer program, when the computer program is executed by processor, execute shown in above-mentioned any one of first aspect
Personage's clothing collocation evaluation method.
Evaluation method, device and the server of personage's clothing collocation provided in an embodiment of the present invention, to the clothing matched
When object is evaluated, the body information and present cloth information of target object are first obtained；And target pair is determined according to body information
The body type of elephant；And then the body type of target object and present cloth information input to clothing are arranged in pairs or groups evaluation model
In, the evaluation result of target object clothing collocation is obtained, aesthetic carries out evaluation phase with the subjectivity for passing through estimator in the prior art
Than, it not will receive the subjective aesthetic influence of estimator, but the clothing by pre-establishing is arranged in pairs or groups, evaluation model is evaluated, from
And improve the accuracy of clothing collocation evaluation.
Detailed description of the invention
The drawings herein are incorporated into the specification and forms part of this specification, and shows the implementation for meeting the disclosure
Example, and together with specification for explaining the principles of this disclosure.
Fig. 1 is a kind of flow diagram of the evaluation method of personage's clothing collocation provided in an embodiment of the present invention；
Fig. 2 is the flow diagram of the evaluation method of another personage's clothing collocation provided in an embodiment of the present invention；
Fig. 3 is the flow diagram of the evaluation method of another personage's clothing provided in an embodiment of the present invention collocation；
Fig. 4 is a kind of structural schematic diagram of the evaluating apparatus of personage's clothing collocation provided in an embodiment of the present invention；
Fig. 5 is the structural schematic diagram of the evaluating apparatus of another personage's clothing collocation provided in an embodiment of the present invention；
Fig. 6 is a kind of structural schematic diagram of server provided in an embodiment of the present invention.
Through the above attached drawings, it has been shown that the specific embodiment of the disclosure will be hereinafter described in more detail.These attached drawings
It is not intended to limit the scope of this disclosure concept by any means with verbal description, but is by referring to specific embodiments
Those skilled in the art illustrate the concept of the disclosure.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Description and claims of this specification and term " first ", " second ", " third " and " in above-mentioned attached drawing
The (if present)s such as four " are to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should manage
The data that solution uses in this way are interchangeable under appropriate circumstances, so as to the embodiment of the present invention described herein can in addition to
Here the sequence other than those of diagram or description is implemented.In addition, term " includes " and " having " and their any deformation,
Be intended to cover it is non-exclusive include, for example, containing the process, method of a series of steps or units, system, product or setting
It is standby those of to be not necessarily limited to be clearly listed step or unit, but may include be not clearly listed or for these mistakes
The intrinsic other step or units of journey, method, product or equipment.
In the prior art, when evaluating the clothing matched, be according to estimator subjectivity it is aesthetic evaluate,
But due to the aesthetic difference of subjectivity of each object, cause the accuracy of clothing collocation evaluation not high.It arranges in pairs or groups and evaluates for clothing
Accuracy, the embodiment of the invention provides a kind of personage's clothing collocation evaluation method, commented to the clothing matched
When valence, the body information and present cloth information of target object are first obtained；And the body of target object is determined according to body information
Type；And then the body type of target object and present cloth information input are obtained into mesh into clothing collocation evaluation model
The evaluation result for marking the collocation of object clothing, compared with being evaluated in the prior art by the way that the subjectivity of estimator is aesthetic, Bu Huishou
To the subjective aesthetic influence of estimator, but the clothing by pre-establishing is arranged in pairs or groups, evaluation model is evaluated, to improve
The accuracy of clothing collocation evaluation.
How technical solution of the present invention and technical solution of the present invention are solved with specific embodiment below above-mentioned
Technical problem is described in detail.These specific embodiments can be combined with each other below, for the same or similar concept
Or process repeats no more in certain embodiments.Below in conjunction with attached drawing, the embodiment of the present invention is described.
Fig. 1 is a kind of flow diagram of the evaluation method of personage's clothing collocation provided in an embodiment of the present invention, the personage
The evaluation method of clothing collocation can be applied to augmented reality, and the evaluation method of personage's clothing collocation can be taken by personage's clothing
The evaluating apparatus matched executes, and the evaluating apparatus of personage's clothing collocation can be independently arranged, and also can integrate in the server.Show
Example, the evaluation method of personage's clothing collocation may include:
S101, the body information and present cloth information for obtaining target object.
Wherein, the body information of target object can refer to that the form of the target object, present cloth information refer to the target pair
As the clothing worn with current.
It is exemplary, in embodiments of the present invention, the picture of target object can be obtained in advance by terminal, and pass through feature
Extractive technique extracts the body information and its present cloth information of the target object, it is of course also possible to be obtained in real time by terminal
The picture of target object, and by Feature Extraction Technology extract the target object body information and its present cloth information,
This, does not do and further limits for the body information and present cloth information, the embodiment of the present invention that how to obtain target object.
S102, the body type that target object is determined according to body information.
Exemplary, body type can be wide for upper body, and the lower part of the body is narrow or upper body is narrow, and the lower part of the body is wide；It is of course also possible to
It is well-balanced etc. for upper body and the lower part of the body.
After the body information for getting target object by above-mentioned S102, so that it may according to the body of the target formation
Information determines the body type of the object of the target.
In S103, evaluation model that the body type of target object and present cloth information input to clothing are arranged in pairs or groups, obtain
The evaluation result of target object clothing collocation.
It is exemplary, when exporting the evaluation result of target object clothing collocation, it can directly export the evaluation of clothing collocation
Resulting class, as arranging in pairs or groups poor, the clothing that excessively poor, the clothing of clothing collocation are arranged in pairs or groups, preferable, the clothing that clothing is arranged in pairs or groups
Object is arranged in pairs or groups very good etc..It is of course also possible to export the quantized value of the evaluation result of clothing collocation, the evaluation knot that clothing is arranged in pairs or groups
Fruit is quantified as score section, for example, score section 0-20 points correspond to the excessively poor of clothings collocation, score section 20-40 points of corresponding clothings are taken
That matches is poor, general, preferable, the score of score section 60-80 points of corresponding clothing collocation of score section 40-60 points of corresponding clothing collocation
It is very good that 80-100 points of section corresponding clothings are arranged in pairs or groups.Here, the embodiment of the present invention is only carried out by taking both way of outputs as an example
Illustrate, but does not represent the embodiment of the present invention and be limited only to this.
After the body type and present cloth information for getting target object respectively, so that it may by the shape of target object
Into clothing collocation evaluation model, the output result of clothing collocation evaluation model is for body type and present cloth information input
The evaluation result of target object clothing collocation, to obtain the evaluation result of target object clothing collocation, and in the prior art
It carries out evaluation by the way that the subjectivity of estimator is aesthetic and compares, not will receive the subjective aesthetic influence of estimator, but by building in advance
Vertical clothing collocation evaluation model is evaluated, to improve the accuracy of clothing collocation evaluation.
The evaluation method of personage's clothing collocation provided in an embodiment of the present invention, when evaluating the clothing matched,
First obtain the body information and present cloth information of target object；And the body type of target object is determined according to body information；
And then the body type of target object and present cloth information input are obtained into target pair into clothing collocation evaluation model
As the evaluation result that clothing is arranged in pairs or groups not will receive and comment compared with evaluating in the prior art by the way that the subjectivity of estimator is aesthetic
The subjective aesthetic influence of valence person, but the clothing by pre-establishing is arranged in pairs or groups, evaluation model is evaluated, to improve clothing
The accuracy of collocation evaluation.
Based on embodiment shown in FIG. 1, for personage's clothing collocation shown in the clearer description embodiment of the present invention
Evaluation method, exemplary, shown in Figure 2, Fig. 2 is the evaluation of another personage's clothing collocation provided in an embodiment of the present invention
The flow diagram of method, the evaluation method which arranges in pairs or groups can also include:
S201, according to the body type in training sample each in multiple training samples, construct each sample corresponding
One feature vector.
Exemplary, body type can be wide for upper body, and the lower part of the body is narrow or upper body is narrow, and the lower part of the body is wide；It is of course also possible to
It is well-balanced etc. for upper body and the lower part of the body.
In embodiments of the present invention, multiple training samples can first be determined, the quantity of the training sample is not particularly limited,
Certainly, the quantity of the training sample of selection is more, and the accuracy for the clothing collocation evaluation model that final training obtains is higher.Its
In, it include body type in each training sample, after getting the body type in each training sample, so that it may
The corresponding first eigenvector of the training sample is constructed according to the body type in each training sample with building.
S202, according to the corresponding first eigenvector of each sample and the clothing information marked in each sample, determine
Clothing collocation evaluation model.
It should be noted that being all marked in the training sample in training sample for each training sample
The corresponding clothing information of body type.For example, the lower part of the body is narrow when the body type of certain an object in a certain training sample is that upper body is wide
When, the corresponding clothing information of the body type can be with are as follows: upper body the collocation not style of shoulder pad and the simple jacket of lines etc., under
Body collocation longuette, a line skirt, pleating skirt etc.；Opposite, when body type is that upper body is narrow, when the lower part of the body is wide, the body type is corresponding
Clothing information can be with are as follows: upper body is collocated with the eye-catching shirt of shoulder pad, collar, Western-style clothes etc., and the lower part of the body is arranged in pairs or groups pencil skirt or panty girdle
Deng；When body type is upper body and the well-balanced lower part of the body, the corresponding clothing information of the body type can be with are as follows: upper body is arranged in pairs or groups on shortage of money
Clothing (T-shirt, vest, reducing garment) etc., high waist trousers of lower part of the body collocation (shorts, close-fitting bound feet trousers) etc..
After respectively obtaining the corresponding first eigenvector of each sample and the clothing information marked in each sample,
The corresponding first eigenvector of each sample and the clothing marked in each sample can be believed using deep learning method
Breath is trained, so that clothing collocation evaluation model is obtained, it is of course also possible to pass through other methods corresponding to each sample the
One feature vector and the clothing information marked in each sample are trained, to obtain clothing collocation evaluation model, here, this
Inventive embodiments using deep learning method to the corresponding first eigenvector of each sample and in each sample only to get the bid
The clothing information of note is trained, and obtains being illustrated for clothing collocation evaluation model, but do not represent the embodiment of the present invention
It is limited only to this.
It should be noted that not being in each evaluation result for determining the collocation of target object clothing for S201-S202
When, it is required to execute the S201-S202, can be established only when first time determining the evaluation result of target object clothing collocation
Clothing collocation evaluation model.It certainly, can be continuous in order to further improve the accuracy of clothing collocation evaluation model
Ground is trained revision to clothing collocation evaluation model, to improve the accuracy of clothing collocation evaluation model.
S203, according to the body information in training sample each in multiple training samples, construct each sample corresponding
Two feature vectors.
Likewise, in embodiments of the present invention, can first determine multiple training samples, the quantity of the training sample, which is not done, to be had
Body limitation, certainly, the quantity of the training sample of selection is more, and the accuracy for the body type detection model that final training obtains is just
It is higher.Wherein, in each training sample include body information, get the body information in each training sample it
Afterwards, so that it may which building is according to the corresponding second feature vector of the body information architecture training sample in each training sample.
S204, the body type marked in each sample according to the corresponding second feature vector sum of each sample are determined
Body type detection model.
After respectively obtaining the body type that the corresponding second feature vector sum of each sample marks in each sample,
The body class that the corresponding second feature vector sum of each sample can be marked in each sample using deep learning method
Type is trained, so that body type detection model is obtained, it is of course also possible to pass through other methods corresponding to each sample
Two feature vectors and the body type marked in each sample are trained, to obtain body type detection model, here, this
Inventive embodiments are only to get the bid to the corresponding second feature vector sum of each sample in each sample using deep learning method
The body type of note is trained, and obtains being illustrated for body type detection model, but do not represent the embodiment of the present invention
It is limited only to this.
It should be noted that for S203-S204 being required in each body type for determining target object
The S203-S204 is executed, the body type detection model can be established only when first time determining the body type of target object
?.It certainly, can be constantly to the body type detection in order to further improve the accuracy of body type detection model
Model is trained revision, to improve the accuracy of the body type detection model.
After getting clothing collocation evaluation model and body type detection model respectively, so that it may be taken by the clothing
Determine the evaluation result of target object clothing collocation jointly with evaluation model and body type detection model.Certainly, mesh is being determined
Before the evaluation result for marking the collocation of object clothing, needs first to obtain the body information and present cloth information of target object, that is, hold
The following S205 of row:
S205, the body information and present cloth information for obtaining target object.
Wherein, the body information of target object can refer to that the form of the target object, present cloth information refer to the target pair
As the clothing worn with current.
Likewise, in embodiments of the present invention, the picture of target object can be obtained in advance by terminal, and pass through feature
Extractive technique extracts the body information and its present cloth information of the target object, it is of course also possible to be obtained in real time by terminal
The picture of target object, and by Feature Extraction Technology extract the target object body information and its present cloth information,
This, does not do and further limits for the body information and present cloth information, the embodiment of the present invention that how to obtain target object.
It is exemplary, when target object is Zhang San, the body information of Zhang San and current can be obtained by the picture of Zhang San
Clothing information, Zhang San's present cloth information are as follows: upper body has been arranged in pairs or groups the shirt for having shoulder pad, collar eye-catching, and the lower part of the body has arranged in pairs or groups one
A longuette.
It should be noted that in embodiments of the present invention, sequencing, Ke Yixian are had no between S201-S204 and S205
S201-S204 is executed, then executes S205；S205 can also be first carried out, then executes S201-S204；It is of course also possible to be performed simultaneously
S201-S204 and S205 here, the embodiment of the present invention is only to first carry out S201-S204, then executed and is said for S205
It is bright, but do not represent the embodiment of the present invention and be limited only to this.
S206, by the body information input of target object into body type detection model, obtain the body of target object
Type.
Exemplary, body type can be wide for upper body, and the lower part of the body is narrow or upper body is narrow, and the lower part of the body is wide；It is of course also possible to
It is well-balanced etc. for upper body and the lower part of the body.
After the body information for getting target object by above-mentioned S205, so that it may believe the body of the target object
Breath is input in body type detection model, to obtain the body type of target object.
It is exemplary, after getting the body information of Zhang San, so that it may by the body information input of Zhang San to building in advance
In vertical body type detection model, thus determine that the body type of Zhang San is that upper body is wide according to the body type detection model,
The lower part of the body is narrow.
In S207, evaluation model that the body type of target object and present cloth information input to clothing are arranged in pairs or groups, obtain
The evaluation result of target object clothing collocation.
Likewise, can directly export the evaluation of clothing collocation when exporting the evaluation result of target object clothing collocation
Resulting class, as arranging in pairs or groups poor, the clothing that excessively poor, the clothing of clothing collocation are arranged in pairs or groups, preferable, the clothing that clothing is arranged in pairs or groups
Object is arranged in pairs or groups very good etc..It is of course also possible to export the quantized value of the evaluation result of clothing collocation, the evaluation knot that clothing is arranged in pairs or groups
Fruit is quantified as score section, for example, score section 0-20 points correspond to the excessively poor of clothings collocation, score section 20-40 points of corresponding clothings are taken
That matches is poor, general, preferable, the score of score section 60-80 points of corresponding clothing collocation of score section 40-60 points of corresponding clothing collocation
It is very good that 80-100 points of section corresponding clothings are arranged in pairs or groups.Here, the embodiment of the present invention is only carried out by taking both way of outputs as an example
Illustrate, but does not represent the embodiment of the present invention and be limited only to this.
After the body type and present cloth information for getting target object respectively, so that it may by the shape of target object
Into clothing collocation evaluation model, the output result of clothing collocation evaluation model is for body type and present cloth information input
The evaluation result of target object clothing collocation, to obtain the evaluation result of target object clothing collocation, and in the prior art
It carries out evaluation by the way that the subjectivity of estimator is aesthetic and compares, not will receive the subjective aesthetic influence of estimator, but by building in advance
Vertical clothing collocation evaluation model is evaluated, to improve the accuracy of clothing collocation evaluation.
It is exemplary, respectively determine Zhang San body type (upper body is wide, and the lower part of the body is narrow), and Zhang San's present cloth information (on
Body has been arranged in pairs or groups the shirt for having shoulder pad, collar eye-catching, and the lower part of the body has been arranged in pairs or groups a longuette) after, so that it may by body type (on
Body is wide, and the lower part of the body is narrow) and present cloth information (upper body has been arranged in pairs or groups the shirt for having shoulder pad, collar eye-catching, and the lower part of the body has been arranged in pairs or groups one
Longuette) corresponding feature vector is input in clothing collocation evaluation model, so that it may determine that the clothing collocation of Zhang San is poor, and it is existing
There is the subjectivity in technology by estimator is aesthetic to carry out evaluation and compare, not will receive the subjective aesthetic influence of estimator, but it is logical
It crosses the clothing collocation evaluation model pre-established to be evaluated, to improve the accuracy of clothing collocation evaluation.
Based on embodiment shown in fig. 1 or fig. 2, optionally, by S207 by the body type of target object and current
Clothing information input is into clothing collocation evaluation model, after obtaining the evaluation result of target object clothing collocation, can also incite somebody to action
The evaluation result of target object clothing collocation is sent to terminal, so that the target object can check the clothing collocation evaluation of oneself
As a result.Further, it is also possible to for the body type according to target object, corresponding clothing collocation strategy is determined, and by the clothing
Object collocation strategy is sent to the target object, so that the target object adjusts current clothing according to clothing collocation strategy
It is whole.Exemplary, shown in Figure 3, Fig. 3 is the evaluation method of another personage's clothing provided in an embodiment of the present invention collocation
Flow diagram, the evaluation method which arranges in pairs or groups can also include:
Corresponding relationship between S301, strategy of being arranged in pairs or groups according to the body type and body type and clothing of target object, really
Fixed corresponding clothing collocation strategy.
Optionally, it is that upper body is loose that the corresponding relationship between body type and clothing collocation strategy, which may include: body type,
The lower part of the body is narrow, and corresponding clothing collocation strategy can be with are as follows: upper body the collocation not style of shoulder pad and the simple jacket of lines etc., the lower part of the body
Collocation longuette, a line skirt, pleating skirt etc.；Body type is that upper body is narrow, and the lower part of the body is wide, and corresponding clothing collocation strategy can be with are as follows: upper body
It is collocated with the eye-catching shirt of shoulder pad, collar, Western-style clothes etc., the lower part of the body collocation pencil skirt or panty girdle etc.；Body type be upper body and
The lower part of the body is well-balanced, and corresponding clothing collocation strategy can be with are as follows: upper body is arranged in pairs or groups shortage of money jacket (T-shirt, vest, reducing garment) etc., and the lower part of the body is taken
With high waist trousers (shorts, close-fitting bound feet trousers) etc..
It is exemplary, the corresponding relationship between body type and clothing collocation strategy can be stored in advance in server, in this way
In the body type for obtaining target object, so that it may pair between body type according to the pre-stored data and clothing collocation strategy
It should be related to, determine the corresponding clothing collocation strategy of the body type of the target object.
It is exemplary, it, can be according to the pre-stored data when the clothing for determining Zhang San by above-mentioned S201-S207 arranges in pairs or groups poor
Corresponding relationship between body type and clothing collocation strategy determines the corresponding clothing collocation strategy of the body type of Zhang San.By
It is that upper body is wide in the body type of Zhang San, the lower part of the body is narrow, in order to avoid the wide defect of prominent upper body, the adjustable current upper body of Zhang San
Collocation, corresponding clothing adjustable strategies can for by the collocation of upper body be adjusted to the not style of shoulder pad and lines it is simple on
Clothing.
S302, the body type for sending evaluation result and target object that target object clothing is arranged in pairs or groups to terminal are corresponding
Clothing collocation strategy.
In the evaluation result clothing corresponding with the body type of the target object for determining the collocation of target object clothing respectively
After collocation strategy, so that it may which evaluation result clothing collocation strategy corresponding with its that target object clothing is arranged in pairs or groups is sent to end
End allows terminal to show the target object clothing using the mode of augmented reality (Augmented Reality, AR)
The evaluation result of collocation, so that the target object can check the evaluation result of the clothing collocation of oneself in time by the terminal,
And collocation is adjusted according to corresponding clothing collocation strategy, to improve the collocation level of wearing the clothes of oneself.
Exemplary, when the clothing collocation for determining Zhang San respectively is poor and the corresponding clothing of body type of Zhang San is arranged in pairs or groups
After strategy, so that it may evaluation result clothing collocation strategy corresponding with its that the clothing of Zhang San is arranged in pairs or groups is sent to terminal, with
Zhang San is set to check the evaluation result of the clothing collocation of oneself in time by the terminal, and according to corresponding clothing collocation strategy
Clothing adjustment is carried out, to improve the collocation level of wearing the clothes of oneself.
Fig. 4 is a kind of structural schematic diagram of the evaluating apparatus 40 of personage's clothing collocation provided in an embodiment of the present invention, please be joined
As shown in Figure 4, the evaluating apparatus 40 of personage's clothing collocation can be applied to augmented reality, the evaluation dress of personage's clothing collocation
Setting 40 may include:
Acquiring unit 401, for obtaining the body information and present cloth information of target object.
Processing unit 402, for determining the body type of target object according to body information.
Processing unit 402 is also used to comment the body type of target object and present cloth information input to clothing collocation
In valence model, the evaluation result of target object clothing collocation is obtained.
Optionally, the evaluating apparatus 40 of personage's clothing collocation further includes construction unit 403, exemplary, refers to Fig. 5 institute
Show, Fig. 5 is the structural schematic diagram of the evaluating apparatus 40 of another personage's clothing collocation provided in an embodiment of the present invention.
Clothing collocation evaluation model is trained by the following method: construction unit 403, for according to multiple training samples
In body type in each training sample, construct the corresponding first eigenvector of each sample.
Processing unit 402 is also used to mark according to the corresponding first eigenvector of each sample and in each sample
Clothing information determines clothing collocation evaluation model.
Optionally, processing unit 402, be specifically used for using deep learning method to the corresponding fisrt feature of each sample to
Amount and the clothing information marked in each sample are trained, and obtain clothing collocation evaluation model.
Optionally, processing unit 402 are specifically used for obtaining body type detection model；And the body of target object is believed
Breath is input in body type detection model, obtains the body type of target object.
Optionally, construction unit 403 are also used to according to the body information in training sample each in multiple training samples,
Construct the corresponding second feature vector of each sample.
Processing unit 402 is also used to be marked in each sample according to the corresponding second feature vector sum of each sample
Body type determines body type detection model.
Optionally, processing unit 402, be specifically used for using deep learning method to the corresponding second feature of each sample to
Amount and the body type marked in each sample are trained, and obtain body type detection model.
Optionally, the evaluating apparatus 40 of personage's clothing collocation further includes transmission unit 404.
Processing unit 402 is also used to according between the body type and body type and clothing of target object collocation strategy
Corresponding relationship, determine corresponding clothing collocation strategy.
Transmission unit 404, for sending the body of evaluation result and target object that target object clothing is arranged in pairs or groups to terminal
The corresponding clothing collocation strategy of type.
The evaluating apparatus 40 of the collocation of personage's clothing shown in the embodiment of the present invention, can execute shown in any of the above-described embodiment
Personage's clothing collocation evaluation method technical solution, realization principle and beneficial effect are similar, no longer go to live in the household of one's in-laws on getting married herein
It states.
Fig. 6 is a kind of structural schematic diagram of server 60 provided in an embodiment of the present invention, shown in Figure 6, the service
Device 60 may include processor 601 and memory 602.Wherein,
Memory 602 is for storing program instruction.
Processor 601 is used to read the program instruction in memory 602, and is held according to the program instruction in memory 602
The evaluation method of personage's clothing collocation shown in any of the above-described embodiment of row.
Terminal device 60 shown in the embodiment of the present invention can execute the collocation of personage's clothing shown in any of the above-described embodiment
Evaluation method technical solution, realization principle and beneficial effect are similar, are no longer repeated herein.
The embodiment of the present invention also provides a kind of computer readable storage medium, is stored on the computer readable storage medium
Computer program executes the collocation of personage's clothing shown in any of the above-described embodiment when computer program is executed by processor
Evaluation method, realization principle and beneficial effect are similar, are no longer repeated herein.
Processor can be general processor, digital signal processor (digital signal in above-described embodiment
Processor, DSP), it is specific integrated circuit (application specific integrated circuit, ASIC), existing
At programmable gate array (field programmable gate array, FPGA) or other programmable logic device, discrete
Door or transistor logic, discrete hardware components.It may be implemented or execute the disclosed each side in the embodiment of the present invention
Method, step and logic diagram.General processor can be microprocessor or the processor is also possible to any conventional processing
Device etc..The step of method in conjunction with disclosed in the embodiment of the present invention, can be embodied directly in hardware decoding processor and execute completion,
Or in decoding processor hardware and software module combination execute completion.Software module can be located at random access memory
(random access memory, RAM), flash memory, may be programmed read-only deposit at read-only memory (read-only memory, ROM)
In the storage medium of this fields such as reservoir or electrically erasable programmable memory, register maturation.The storage medium, which is located at, to be deposited
The step of reservoir, processor reads the instruction in memory, completes the above method in conjunction with its hardware.
In several embodiments provided by the present invention, it should be understood that disclosed device and method can pass through it
Its mode is realized.For example, the apparatus embodiments described above are merely exemplary, for example, the division of the unit, only
Only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components can be tied
Another system is closed or is desirably integrated into, or some features can be ignored or not executed.Another point, it is shown or discussed
Mutual coupling, direct-coupling or communication connection can be through some interfaces, the INDIRECT COUPLING or logical of device or unit
Letter connection can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.In addition, the functional units in various embodiments of the present invention may be integrated into one processing unit, it is also possible to each
Unit physically exists alone, and can also be integrated in one unit with two or more units.Above-mentioned integrated unit both may be used
To use formal implementation of hardware, can also be realized in the form of hardware adds SFU software functional unit.
Those skilled in the art after considering the specification and implementing the invention disclosed here, will readily occur to its of the disclosure
Its embodiment.The present invention is directed to cover any variations, uses, or adaptations of the disclosure, these modifications, purposes or
Person's adaptive change follows the general principles of this disclosure and including the undocumented common knowledge in the art of the disclosure
Or conventional techniques.The description and examples are only to be considered as illustrative, and the true scope and spirit of the disclosure are by following
Claims are pointed out.
It should be understood that the present disclosure is not limited to the precise structures that have been described above and shown in the drawings, and
And various modifications and changes may be made without departing from the scope thereof.The scope of the present disclosure is only limited by appended claims
System.
Claims (16)
1. a kind of evaluation method of personage's clothing collocation, which is characterized in that the method is used for augmented reality, the method packet
It includes:
Obtain the body information and present cloth information of target object；
The body type of the target object is determined according to the body information；
By the body type of the target object and the present cloth information input into clothing collocation evaluation model, institute is obtained
State the evaluation result of target object clothing collocation.
2. the method according to claim 1, wherein clothing collocation evaluation model is to instruct by the following method
Experienced:
According to the body type in training sample each in multiple training samples, construct the corresponding fisrt feature of each sample to
Amount；
According to the corresponding first eigenvector of each sample and the clothing information marked in each sample, determine that the clothing is taken
With evaluation model.
3. according to the method described in claim 2, it is characterized in that, it is described according to the corresponding first eigenvector of each sample and
The clothing information marked in each sample determines the clothing collocation evaluation model, comprising:
Using deep learning method to the corresponding first eigenvector of each sample and the clothing information marked in each sample
It is trained, obtains the clothing collocation evaluation model.
4. the method according to claim 1, wherein described determine the target object according to the body information
Body type, comprising:
Obtain body type detection model；
By the body information input of the target object into the body type detection model, the shape of the target object is obtained
Body type.
5. according to the method described in claim 4, it is characterized in that, the acquisition body type detection model, comprising:
According to the body information in training sample each in multiple training samples, construct the corresponding second feature of each sample to
Amount；
According to the body type that the corresponding second feature vector sum of each sample marks in each sample, the body class is determined
Type detection model.
6. according to the method described in claim 5, it is characterized in that, described according to the corresponding second feature vector sum of each sample
The body type marked in each sample determines the body type detection model, comprising:
The body type that the corresponding second feature vector sum of each sample is marked in each sample using deep learning method
It is trained, obtains the body type detection model.
7. method according to claim 1-6, which is characterized in that the body type by the target object
With the present cloth information input into clothing collocation evaluation model, the evaluation result of the target object clothing collocation is obtained
Later, further includes:
According to the corresponding relationship between the body type of the target object and body type and clothing collocation strategy, phase is determined
The clothing collocation strategy answered；
The corresponding clothing of body type of evaluation result and the target object that the target object clothing is arranged in pairs or groups is sent to terminal
Object collocation strategy.
8. a kind of evaluating apparatus of personage's clothing collocation, which is characterized in that described device is used for augmented reality, described device packet
It includes:
Acquiring unit, for obtaining the body information and present cloth information of target object；
Processing unit, for determining the body type of the target object according to the body information；
The processing unit is also used to take the body type of the target object and the present cloth information input to clothing
With the evaluation result in evaluation model, obtaining the target object clothing collocation.
9. device according to claim 8, which is characterized in that described device further includes construction unit, the clothing collocation
Evaluation model is trained by the following method:
The construction unit, for constructing each sample according to the body type in training sample each in multiple training samples
Corresponding first eigenvector；
The processing unit is also used to according to the clothing each sample corresponding first eigenvector and marked in each sample
Information determines the clothing collocation evaluation model.
10. device according to claim 9, which is characterized in that
The processing unit is specifically used for using deep learning method to the corresponding first eigenvector of each sample and each
The clothing information marked in sample is trained, and obtains the clothing collocation evaluation model.
11. device according to claim 8, which is characterized in that
The processing unit is specifically used for obtaining body type detection model；And by the body information input of the target object
To in the body type detection model, the body type of the target object is obtained.
12. device according to claim 11, which is characterized in that
Construction unit is also used to construct each sample pair according to the body information in training sample each in multiple training samples
The second feature vector answered；
The processing unit is also used to the body marked in each sample according to the corresponding second feature vector sum of each sample
Type determines the body type detection model.
13. device according to claim 11, which is characterized in that
The processing unit is specifically used for using deep learning method to the corresponding second feature vector sum of each sample each
The body type marked in sample is trained, and obtains the body type detection model.
14. according to the described in any item devices of claim 8-13, which is characterized in that further include transmission unit；
The processing unit is also used to strategy of arranging in pairs or groups according to the body type and body type and clothing of the target object
Between corresponding relationship, determine corresponding clothing collocation strategy；
The transmission unit, for sending the evaluation result and the target object that the target object clothing is arranged in pairs or groups to terminal
Body type corresponding clothing collocation strategy.
15. a kind of server, which is characterized in that including processor and memory, wherein
The memory is for storing program instruction；
The processor is used to read the program instruction in the memory, and is executed according to the program instruction in the memory
The evaluation method of the collocation of personage's clothing shown in any one of claim 1~7.
16. a kind of computer readable storage medium, which is characterized in that
It is stored with computer program on computer readable storage medium, when the computer program is executed by processor, executes
The evaluation method of the collocation of personage's clothing shown in any one of claim 1~7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810887103.2A CN108985657A (en) | 2018-08-06 | 2018-08-06 | Evaluation method, device and the server of personage's clothing collocation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810887103.2A CN108985657A (en) | 2018-08-06 | 2018-08-06 | Evaluation method, device and the server of personage's clothing collocation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108985657A true CN108985657A (en) | 2018-12-11 |
Family
ID=64555236
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810887103.2A Pending CN108985657A (en) | 2018-08-06 | 2018-08-06 | Evaluation method, device and the server of personage's clothing collocation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108985657A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101477560A (en) * | 2009-02-02 | 2009-07-08 | 林�智 | Clothing matching system and method based on scheme database |
| CN103886115A (en) * | 2012-12-20 | 2014-06-25 | 上海工程技术大学 | Building and calling method of three-dimension virtual body form based on different body types |
| CN104484816A (en) * | 2014-12-19 | 2015-04-01 | 常州飞寻视讯信息科技有限公司 | Clothing shopping guide system and clothing shopping guide method based on big data analysis |
| CN105809461A (en) * | 2014-12-30 | 2016-07-27 | Tcl集团股份有限公司 | Method, system and intelligent device for recommending jewelry |
| US20170083789A1 (en) * | 2015-09-22 | 2017-03-23 | Swati Shah | Clothing matching system and method |
| CN106709781A (en) * | 2016-12-05 | 2017-05-24 | 姚震亚 | Personal image design and collocation purchasing device and method |
| CN107292685A (en) * | 2016-03-30 | 2017-10-24 | 深圳市祈飞科技有限公司 | A kind of method of automatic recommendation size and the fitting cabinet system using this method |
-
2018
- 2018-08-06 CN CN201810887103.2A patent/CN108985657A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101477560A (en) * | 2009-02-02 | 2009-07-08 | 林�智 | Clothing matching system and method based on scheme database |
| CN103886115A (en) * | 2012-12-20 | 2014-06-25 | 上海工程技术大学 | Building and calling method of three-dimension virtual body form based on different body types |
| CN104484816A (en) * | 2014-12-19 | 2015-04-01 | 常州飞寻视讯信息科技有限公司 | Clothing shopping guide system and clothing shopping guide method based on big data analysis |
| CN105809461A (en) * | 2014-12-30 | 2016-07-27 | Tcl集团股份有限公司 | Method, system and intelligent device for recommending jewelry |
| US20170083789A1 (en) * | 2015-09-22 | 2017-03-23 | Swati Shah | Clothing matching system and method |
| CN107292685A (en) * | 2016-03-30 | 2017-10-24 | 深圳市祈飞科技有限公司 | A kind of method of automatic recommendation size and the fitting cabinet system using this method |
| CN106709781A (en) * | 2016-12-05 | 2017-05-24 | 姚震亚 | Personal image design and collocation purchasing device and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020052169A1 (en) | Clothing attribute recognition detection method and apparatus | |
| CN104952113B (en) | Dress ornament tries experiential method, system and equipment on | |
| CN105657666B (en) | A kind of commercial employment crowd residence recognition methods based on mobile phone location data | |
| CN105224775A (en) | Based on the method and apparatus that picture processing is arranged in pairs or groups to clothes | |
| CN107340852A (en) | Gestural control method, device and terminal device | |
| CN110264299A (en) | Clothes recommended method, device and computer equipment based on recognition of face | |
| CN109766872A (en) | Image-recognizing method and device | |
| CN105938564A (en) | Rice disease recognition method based on principal component analysis and neural network and rice disease recognition system thereof | |
| CN108205654A (en) | A kind of motion detection method and device based on video | |
| CN103489008B (en) | Image product information extracting method and device | |
| CN107784372A (en) | Forecasting Methodology, the device and system of destination object attribute | |
| CN102013071A (en) | Virtual fitting method based on computer system | |
| CN110069820A (en) | A kind of Design System of Garment of designer | |
| CN107886344A (en) | Convolutional neural network-based cheating advertisement page identification method and device | |
| CN105913297A (en) | Fitting recommendation method based on big data and system | |
| CN108876913A (en) | A kind of intelligent clothing clip system and method | |
| CN107729380A (en) | Clothing matching method, terminal, terminal | |
| CN110069983A (en) | Vivo identification method, device, terminal and readable medium based on display medium | |
| CN109543453A (en) | A kind of order information processing method based on multi-party computations, apparatus and system | |
| CN110162454A (en) | Game running method and device, storage medium and electronic device | |
| CN109992781A (en) | Processing, device, storage medium and the processor of text feature | |
| CN108108897A (en) | A kind of track traffic for passenger flow classifying method, system and electronic equipment | |
| CN103399900B (en) | Picture recommendation method based on location-based service | |
| CN109978720A (en) | Wear methods of marking, device, smart machine and storage medium | |
| CN108985657A (en) | Evaluation method, device and the server of personage's clothing collocation |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181211 |