CN105488470A - Method and apparatus for determining character attribute information - Google Patents
Method and apparatus for determining character attribute information Download PDFInfo
- Publication number
- CN105488470A CN105488470A CN201510857441.8A CN201510857441A CN105488470A CN 105488470 A CN105488470 A CN 105488470A CN 201510857441 A CN201510857441 A CN 201510857441A CN 105488470 A CN105488470 A CN 105488470A
- Authority
- CN
- China
- Prior art keywords
- picture
- pictures
- sex
- attribute information
- personage
- 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
- G06V40/161—Detection; Localisation; Normalisation
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention relates to a method and an apparatus for determining character attribute information, and belongs to the technical field of human face identification. The method comprises: obtaining a plurality of pictures containing human faces of character objects; performing feature identification on the human faces of the character objects contained in the pictures; determining character attribute information corresponding to the pictures; in combination with the character attribute information corresponding to the pictures, determining character attribute information of the character objects; identifying different human face pictures of the same character object; and synthesizing identification results to determine final character attribute information. Therefore, the effect of improving the accuracy of character attribute information identification is achieved.
Description
Technical field
The disclosure relates to technical field of face recognition, particularly a kind of method and device determining personage's attribute information.
Background technology
Personage's attribute information is identified in a lot of field such as image procossing, authentication and robot vision and all there is potential application.
In the related, from facial image, usually extract the geometric properties required for the identification of personage's attribute information, and use various machine learning algorithm to train the geometric properties extracted.This technology is subject to the impact of the brightness of human face region and lighting angle comparatively large, and the geometric properties extracted under dark or backlight environment has bigger difference relative under home.
Summary of the invention
Disclosure embodiment provides a kind of method and the device of determining personage's attribute information, and technical scheme is as follows:
According to the first aspect of disclosure embodiment, provide a kind of method determining personage's attribute information, the method comprises:
Obtain the picture that multiple comprise the face of same who object;
Respectively feature identification is carried out to the face of the who object comprised in plurality of pictures, determines each self-corresponding personage's attribute information of plurality of pictures; Personage's attribute information comprises age and/or sex;
In conjunction with each self-corresponding personage's attribute information of plurality of pictures, determine personage's attribute information of who object.
Alternatively, respectively feature identification is carried out to the face of the who object comprised in plurality of pictures, determines each self-corresponding personage's attribute information in plurality of pictures, comprising:
When personage's attribute information comprises the age, for each pictures in plurality of pictures, obtain the shooting time of picture;
Feature identification is carried out to the face of the who object comprised in picture, the age of who object when determining that picture is taken;
Age of who object and current time when being taken according to shooting time, picture, determine picture corresponding, age that character image is current;
By corresponding for picture, that character image is current age, be defined as the age that picture is corresponding.
Alternatively, in conjunction with each self-corresponding personage's attribute information of plurality of pictures, determine personage's attribute information of who object, comprising:
When personage's attribute information comprises the age, plurality of pictures each self-corresponding age is averaged, obtain the age of who object.
Alternatively, in conjunction with each self-corresponding personage's attribute information of plurality of pictures, determine personage's attribute information of who object, comprising:
When personage's attribute information comprises sex, determine a sex of large percentage shared in each self-corresponding sex of plurality of pictures;
A sex of large percentage shared in each self-corresponding sex of plurality of pictures is defined as the sex of who object.
Alternatively, the method also comprises:
Before a sex of large percentage shared in each self-corresponding sex of plurality of pictures is defined as the sex of who object, whether the ratio that a sex of the large percentage shared by detecting in each self-corresponding sex of plurality of pictures is corresponding is greater than predetermined threshold value;
If testing result is ratio be greater than predetermined threshold value, then perform a step sex of large percentage shared in each self-corresponding sex of plurality of pictures being defined as the sex of who object.
Alternatively, obtain the picture that multiple comprise the face of same who object, comprising:
According to user account corresponding, the quantity determination numerical value n of the picture that comprises face, n is the integer of >=1;
To user account corresponding, each the picture comprising face carry out recognition of face, obtain the picture cluster that each who object is corresponding, each pictures in a picture cluster all comprises the face of who object corresponding to picture cluster;
The quantity order from big to small of the picture cluster of each who object according to the picture comprised is arranged;
Each picture that a picture cluster in front n picture cluster comprises is retrieved as multiple pictures comprising the face of same who object.
According to the second aspect of disclosure embodiment, provide a kind of device determining personage's attribute information, this device comprises:
Picture acquisition module, is configured to for obtaining the picture that multiple comprise the face of same who object;
First determination module, is configured to for carrying out feature identification respectively to the face of the who object comprised in plurality of pictures, determines each self-corresponding personage's attribute information of plurality of pictures; Personage's attribute information comprises age and/or sex;
Second determination module, is configured to, in conjunction with each self-corresponding personage's attribute information of plurality of pictures, determine personage's attribute information of who object.
Alternatively, the first determination module, comprising:
Time obtains submodule, is configured to for when personage's attribute information comprises the age, for each pictures in plurality of pictures, obtains the shooting time of picture;
First age determined submodule, was configured to for carrying out feature identification to the face of the who object comprised in picture, the age of who object when determining that picture is taken;
Second age determined submodule, age of who object and current time when being configured to for being taken according to shooting time, picture, determine picture corresponding, age that character image is current;
3rd age determined submodule, was configured to, for by corresponding for picture, that character image is current age, be defined as the age that picture is corresponding.
Alternatively, the second determination module, comprising:
Mean value calculation submodule, is configured to, for when personage's attribute information comprises the age, average to plurality of pictures each self-corresponding age, obtains the age of who object.
Alternatively, the second determination module, comprising:
First sex determination submodule, is configured to for when personage's attribute information comprises sex, determines a sex of large percentage shared in each self-corresponding sex of plurality of pictures;
Second sex determination submodule, is configured to the sex for a sex of large percentage shared in each self-corresponding sex of plurality of pictures being defined as who object.
Alternatively, tie the second determination module, also comprise:
Detection sub-module, be configured in before a sex of large percentage shared in each self-corresponding sex of plurality of pictures is defined as the sex of who object by second sex determination submodule, whether the ratio detecting a sex of large percentage shared in each self-corresponding sex of plurality of pictures corresponding is greater than predetermined threshold value;
Second sex determination submodule, the testing result being configured in detection sub-module is that ratio is greater than predetermined threshold value, performs a step sex of large percentage shared in each self-corresponding sex of plurality of pictures being defined as the sex of who object.
Alternatively, picture acquisition module, comprising:
Numerical value determination submodule, be configured to for according to user account corresponding, the quantity determination numerical value n of the picture that comprises face, n is the integer of >=1;
Cluster submodule, be configured to for user account corresponding, each the picture comprising face carry out recognition of face, obtain the picture cluster that each who object is corresponding, each pictures in a picture cluster all comprises the face of who object corresponding to picture cluster;
Sorting sub-module, is configured to for the quantity order from big to small of the picture cluster of each who object according to the picture comprised being arranged;
Picture obtains submodule, and each picture be configured to for a picture cluster in front n picture cluster being comprised is retrieved as multiple pictures comprising the face of same who object.
According to the third aspect of disclosure embodiment, provide a kind of device determining personage's attribute information, this device comprises:
Processor;
For the storer of the executable instruction of storage of processor;
Wherein, processor is configured to:
Obtain the picture that multiple comprise the face of who object;
Respectively feature identification is carried out to the face of the who object comprised in plurality of pictures, determines each self-corresponding personage's attribute information of plurality of pictures; Personage's attribute information comprises age and/or sex;
In conjunction with each self-corresponding personage's attribute information of plurality of pictures, determine personage's attribute information of who object.
The technical scheme that disclosure embodiment provides can comprise following beneficial effect:
By obtaining the picture that multiple comprise the face of who object, respectively feature identification is carried out to the face of the who object comprised in plurality of pictures, determine each self-corresponding personage's attribute information of plurality of pictures, personage's attribute information of this who object is determined again in conjunction with each self-corresponding personage's attribute information of plurality of pictures, the different face picture of same who object are identified respectively, and comprehensively identify that the result of acquisition determines final personage's attribute information respectively, thus reach the effect improving personage's attribute information identification accuracy.
Should be understood that, it is only exemplary and explanatory that above general description and details hereinafter describe, and can not limit the disclosure.
Accompanying drawing explanation
Accompanying drawing to be herein merged in instructions and to form the part of this instructions, shows and meets embodiment of the present disclosure, and is used from instructions one and explains principle of the present disclosure.
Fig. 1 is the schematic diagram of the implementation environment involved by a kind of method determining personage's attribute information according to an exemplary embodiment;
Fig. 2 is a kind of process flow diagram determining the method for personage's attribute information according to an exemplary embodiment;
Fig. 3 A is a kind of process flow diagram determining the method for personage's attribute information according to another exemplary embodiment;
Fig. 3 B is a kind of process flow diagram determining the method at who object age in single picture according to Fig. 3 A illustrated embodiment;
Fig. 4 is a kind of process flow diagram determining the method for personage's attribute information Gen Ju an exemplary embodiment again;
Fig. 5 is a kind of block diagram determining the device of personage's attribute information according to an exemplary embodiment;
Fig. 6 is a kind of block diagram determining the device of personage's attribute information according to another exemplary embodiment;
Fig. 7 is the block diagram of a kind of device according to an exemplary embodiment.
Embodiment
Here will be described exemplary embodiment in detail, its sample table shows in the accompanying drawings.When description below relates to accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawing represents same or analogous key element.Embodiment described in following exemplary embodiment does not represent all embodiments consistent with the disclosure.On the contrary, they only with as in appended claims describe in detail, the example of apparatus and method that aspects more of the present disclosure are consistent.
Fig. 1 is the schematic diagram of the implementation environment involved by method of determination personage attribute information shown by the disclosure.This implementation environment can comprise: terminal 101 and server 102.
Terminal 101 can be the equipment such as intelligent movable phone, PC (PersonalComputer, PC), panel computer, kneetop computer or PDA (personal digital assistant, PersonalDigitalAssistant).
Server 102 can be single server, or the service cluster be made up of multiple servers, or, also can be a cloud computing center.
Terminal 101 is connected with server 102 by wired or wireless network.
Below, for the implementation environment shown in Fig. 1, the technical scheme that each embodiment of the disclosure provides is introduced and is illustrated.
Fig. 2 is a kind of process flow diagram determining the method for personage's attribute information according to an exemplary embodiment, and the method can be applied in the terminal 101 of implementation environment as shown in Figure 1 or in server 102.The method of this calculating personage attribute information can comprise following several step:
In step 201, the picture that multiple comprise the face of same who object is obtained.
In step 202., respectively feature identification is carried out to the face of the who object comprised in this plurality of pictures, determine each self-corresponding personage's attribute information of this plurality of pictures; Personage's attribute information comprises age and/or sex.
In step 203, in conjunction with each self-corresponding personage's attribute information of this plurality of pictures, personage's attribute information of who object is determined.
In sum, the method of the determination personage attribute information that disclosure embodiment provides, by obtaining the picture that multiple comprise the face of who object, respectively feature identification is carried out to the face of the who object comprised in plurality of pictures, determine each self-corresponding personage's attribute information of plurality of pictures, personage's attribute information of this who object is determined again in conjunction with each self-corresponding personage's attribute information of plurality of pictures, the different face picture of same who object are identified respectively, and comprehensively identify that the result of acquisition determines final personage's attribute information respectively, thus reach the effect improving personage's attribute information identification accuracy.
Fig. 3 A is a kind of process flow diagram calculating the method for personage's attribute information according to another exemplary embodiment.Server 102 in this way as shown in Figure 1 in implementation environment performs as example, and please refer to Fig. 3 A, the method can comprise following several step:
In step 301, the picture that multiple comprise the face of same who object is obtained.
When acquisition comprises the picture of the face of same who object, can be read from the photograph album being stored in high in the clouds by server, also by terminal, the picture of storage can be sent to server.Disclosure embodiment does not limit this.
In step 302, respectively feature identification is carried out to the face of the who object comprised in plurality of pictures, determine each self-corresponding personage's attribute information of plurality of pictures; Personage's attribute information comprises age and/or sex.
Such as, the quantity of the picture got is five, and this five pictures all includes the face of who object Jack.Server carries out feature identification respectively to the face parts of images of Jack in this five pictures, and determine personage's attribute information of Jack in each pictures, personage's attribute information can only have age information; Or, only have gender information; Or, simultaneously containing has age and gender information.
In the scheme shown in disclosure embodiment, when personage's attribute information comprises the age, the shooting time in conjunction with picture is needed to determine.Please refer to Fig. 3 B, it illustrates a kind of process flow diagram determining the method at who object age in single picture, the method can comprise the steps:
In step 302a, for each pictures in plurality of pictures, obtain the shooting time of this picture.
When one pictures is taken, capture apparatus can record the shooting time of this photo usually, and is stored in the attribute information of this photo by this shooting time, and server directly can read the attribute information of this photo to obtain its shooting time.
Such as, when personage's attribute information comprises the age, the shooting time of each pictures in five pictures is obtained.Shooting time can be accepted or rejected according to the actual requirements, and which day can be as accurate as, such as on October 5th, 2015; Also can be as accurate as which moon, such as in October, 2015; Or, which year can also be accurate to, such as 2015.
In step 302b, feature identification is carried out to the face of the who object comprised in this picture, the age of this who object when determining that this picture is taken.
Server carries out feature identification to the face of the Jack in a pictures, and direct age of acquisition is exactly the age of this pictures Jack when being taken.
In step 302c, age of who object and current time when being taken according to shooting time, this picture, determine this picture corresponding, age that character image is current.
Such as, a shooting time comprising the picture of Jack facial image is 2013, and when determining that this picture is taken by feature identification, the age of Jack is 28 years old, and current time is 2015.Server current time deducts shooting time 2013 for 2015, obtains the mistiming of 2 years, when picture is taken the age of Jack basis on add mistiming of 2 years, obtain this pictures corresponding, age that Jack is current is 30 years old.
In step 302d, by corresponding for this picture, that character image is current age, be defined as the age that picture is corresponding.
Corresponding for this pictures, that Jack is current age (30 years old) is defined as age corresponding to this pictures by server.
In step 303, when personage's attribute information comprises the age, plurality of pictures each self-corresponding age is averaged, obtain the age of who object.
Such as, when personage's attribute information comprises the age, what who object Jack was corresponding has five pictures, server determines that after feature identification in step 302 age that this 5 pictures is corresponding is 30 years old, 26 years old, 25 years old, 28 years old and 28 years old respectively, the mean value (27.4 years old) at these five ages is defined as the age of Jack.
Optionally, in the disclosed embodiments, in order to ensure the accuracy that the age is determined, can also screen at the age corresponding to each pictures, remove a maximal value and a minimum value, and the mean value at each age remaining is defined as the age of who object.Such as, 30 years old, 26 years old, 25 years old, 28 years old and 28 years old respectively for the age that 5 pictures are corresponding, first server is removed 30 years old and 25 years old these two age, mean value (the 26+28+28)/3=27.3 (year) at other three ages is defined as the age of Jack.
In step 304, when personage's attribute information comprises sex, determine a sex of large percentage shared in each self-corresponding sex of plurality of pictures.
Such as, the quantity of the picture got is five, and this five pictures all includes the face of who object Jack.Server carries out sex identification to this five pictures, recognition result be wherein four for the male sex, one is women, then a sex of server determination large percentage is the male sex, and proportion is 80%.
In step 305, whether the ratio that a sex of the large percentage shared by detecting in each self-corresponding sex of plurality of pictures is corresponding is greater than predetermined threshold value.
Within step 306, if testing result is ratio be greater than predetermined threshold value, then a sex of large percentage shared in each self-corresponding sex of plurality of pictures is defined as the sex of who object.
Wherein, predetermined threshold value can be determined according to the demand of practical application by developer.Such as, predetermined threshold value is 70%, and the sex of large percentage that server is determined in step 305 is the male sex, and the ratio 80% of its correspondence, is greater than predetermined threshold value 70%, then server can determine that Jack is the male sex.
When another possibility, if in five pictures 3 be the male sex, 2 is women, then the ratio shared by any one sex is all less than predetermined threshold value 70%, and now, server can return the result that comprises " None-identified personage sex " information.
In sum, the method for the determination personage attribute information that disclosure embodiment provides, by obtaining the picture that multiple comprise the face of who object; Respectively feature identification is carried out to the face of the who object comprised in plurality of pictures, determine each self-corresponding personage's attribute information of plurality of pictures, personage's attribute information of this who object is determined again in conjunction with each self-corresponding personage's attribute information of plurality of pictures, the different face picture of same who object are identified respectively, and comprehensively identify that the result of acquisition determines final personage's attribute information respectively, thus reach the effect improving personage's attribute information identification accuracy.
In the scheme shown in above-mentioned Fig. 3 A, when acquisition comprises the picture of the face of same who object, obtain the picture that can store from corresponding unique user account.Please refer to Fig. 4, it illustrates a kind of method obtaining picture.Server 102 in this way as shown in Figure 1 in implementation environment performs as example, and please refer to Fig. 4, the method can comprise following several step:
In step 401, to this user account corresponding, each the picture comprising face carry out recognition of face, obtains the picture cluster that each who object is corresponding.
Wherein, each pictures in this picture cluster all comprises the face of who object corresponding to this picture cluster.
Such as, user, can by picture uploading to the server in high in the clouds by after terminal taking picture.The picture same user uploaded of server is to should the account of user store.For each user account, server can carry out recognition of face to each pictures corresponding to this user account, and the picture comprising same who object is carried out cluster.User can check the picture cluster of each who object in end side, and the operation can be correlated with to each picture cluster, such as rename, wherein deletion picture or wherein interpolation picture etc.
In step 402, according to user account corresponding, the quantity determination numerical value n of the picture that comprises face, n is the integer of >=1.
In step 403, the quantity order from big to small of the picture cluster of this each who object according to the picture comprised is arranged.
In step 404, each picture that a picture cluster in front n picture cluster comprises is retrieved as these multiple comprise the picture of the face of same who object.
In order to ensure the accuracy to the identification of personage's attribute information, the picture needs comprising same who object got are abundant, and especially for the identification at age, the picture got is more, and the result of identification is more close to actual result.In the picture that user account is corresponding, usually the picture of who object few in number is only had to meet quantitative requirement, therefore, in scheme shown in disclosure embodiment, for the picture that a user account is corresponding, respectively cluster is carried out to wherein each who object, only to comprise picture number maximum before n picture cluster carry out the identification of personage's attribute information of the who object of its correspondence.This numerical value n can according to this user account corresponding stored, the quantity of the picture that comprises face determines, and numerical value n and this user account corresponding stored, the quantity of the picture that comprises face is directly proportional, such as, when user account corresponding, the picture number that comprises face be 200 time, can determine that numerical value n is 2; When user account corresponding, the picture number that comprises face be 400 time, can determine that numerical value n is 4.
In sum, the method of the acquisition picture that disclosure embodiment provides, by corresponding to this user account, each the picture comprising face carries out recognition of face, obtain the picture cluster that each who object is corresponding, corresponding according to a user account, comprise the quantity determination numerical value n of the picture of face, the quantity order from big to small of the picture cluster of this each who object according to the picture comprised is arranged, and each picture that a picture cluster in front n picture cluster comprises is retrieved as these multiple comprise the picture of the face of same who object, when carrying out the identification of personage's attribute information in units of unique user account, ensure that the picture number comprising the face of same who object got enough meets the requirement of accuracy of identification, thus ensure the accuracy to the identification of personage's attribute information.
It should be noted that, corresponding method shown in above-mentioned Fig. 3 A and Fig. 4, all be that executive agent is illustrated with server, in actual applications, said method also can be performed by the terminal of user, namely terminal carries out cluster to the picture that local or high in the clouds store according to who object, therefrom selects n and comprises the maximum picture cluster of picture number, respectively each picture cluster is carried out to the identification of personage's attribute information of who object.
Fig. 5 is a kind of block diagram determining the device of personage's attribute information according to an exemplary embodiment.This mode determining that the device of personage's attribute information can be combined by hardware or soft or hard is embodied as a part for terminal or server, to perform all or part of step as shown in Fig. 2, Fig. 3 A or Fig. 4 any embodiment.Please refer to Fig. 5, this device can comprise: picture acquisition module 501, first determination module 502 and the second determination module 503.
Picture acquisition module 501, is set to for obtaining the picture that multiple comprise the face of same who object.
First determination module 502, is set to for carrying out feature identification respectively to the face of the who object comprised in plurality of pictures, determines each self-corresponding personage's attribute information of plurality of pictures; Personage's attribute information comprises age and/or sex.
Second determination module 503, is set to, in conjunction with each self-corresponding personage's attribute information of plurality of pictures, determine personage's attribute information of who object.
In sum, the device of the determination personage attribute information that disclosure embodiment provides, by corresponding to this user account, each the picture comprising face carries out recognition of face, obtain the picture cluster that each who object is corresponding, corresponding according to a user account, comprise the quantity determination numerical value n of the picture of face, the quantity order from big to small of the picture cluster of this each who object according to the picture comprised is arranged, and each picture that a picture cluster in front n picture cluster comprises is retrieved as these multiple comprise the picture of the face of same who object, when carrying out the identification of personage's attribute information in units of unique user account, ensure that the picture number comprising the face of same who object got enough meets the requirement of accuracy of identification, thus ensure the accuracy to the identification of personage's attribute information.
Fig. 6 is a kind of block diagram determining the device of personage's attribute information according to another exemplary embodiment.This mode determining that the device of personage's attribute information can be combined by hardware or soft or hard is embodied as a part for terminal or server, to perform all or part of step as shown in Fig. 2, Fig. 3 A or Fig. 4 any embodiment.Please refer to Fig. 6, this device can comprise: picture acquisition module 601, first determination module 602 and the second determination module 603.
Picture acquisition module 601, is set to for obtaining the picture that multiple comprise the face of same who object.
First determination module 602, is set to for carrying out feature identification respectively to the face of the who object comprised in plurality of pictures, determines each self-corresponding personage's attribute information of plurality of pictures; Personage's attribute information comprises age and/or sex.
Second determination module 603, is set to, in conjunction with each self-corresponding personage's attribute information of plurality of pictures, determine personage's attribute information of who object.
Alternatively, the first determination module 602, comprising: the time obtains submodule 602a, and the first age determined submodule 602b, and the second age determined that submodule 602c and the 3rd age determined submodule 602d.
Time obtains submodule 602a, is set to for when personage's attribute information comprises the age, for each pictures in plurality of pictures, obtains the shooting time of picture.
First age determined submodule 602b, was set to for carrying out feature identification to the face of the who object comprised in picture, the age of who object when determining that picture is taken.
Second age determined submodule 602c, age of who object and current time when being set to for being taken according to shooting time, picture, determine picture corresponding, age that character image is current.
3rd age determined submodule 602d, was set to, for by corresponding for picture, that character image is current age, be defined as the age that picture is corresponding.
Alternatively, the second determination module 603, comprising: mean value calculation submodule 603a.
Mean value calculation submodule 603a, is set to, for when personage's attribute information comprises the age, average to plurality of pictures each self-corresponding age, obtains the age of who object.
Second determination module 603, also comprises: the first sex determination submodule 603b and second sex determination submodule 603c.
First sex determination submodule 603b, is set to for when personage's attribute information comprises sex, determines a sex of large percentage shared in each self-corresponding sex of plurality of pictures.
Second sex determination submodule 603c, is set to the sex for a sex of large percentage shared in each self-corresponding sex of plurality of pictures being defined as who object.
Alternatively, the second determination module 603, also comprises: second sex determination submodule 603c and detection sub-module 603d.
Detection sub-module 603d, be arranged in use in before a sex of large percentage shared in each self-corresponding sex of plurality of pictures is defined as the sex of who object by second sex determination submodule 603c, whether the ratio detecting a sex of large percentage shared in each self-corresponding sex of plurality of pictures corresponding is greater than predetermined threshold value.
Second sex determination submodule 603c, the testing result being also arranged in use in detection sub-module 603d is that ratio is greater than predetermined threshold value, performs a step sex of large percentage shared in each self-corresponding sex of plurality of pictures being defined as the sex of who object.
Alternatively, picture acquisition module 601, comprising: numerical value determination submodule 601a, and cluster submodule 601b, sorting sub-module 601c and picture obtain submodule 601d.
Numerical value determination submodule 601a, be set to for according to user account corresponding, the quantity determination numerical value n of the picture that comprises face, n is the integer of >=1.
Cluster submodule 601b, be set to for user account corresponding, each the picture comprising face carry out recognition of face, obtain the picture cluster that each who object is corresponding, each pictures in a picture cluster all comprises the face of who object corresponding to picture cluster.
Sorting sub-module 601c, is set to for the quantity order from big to small of the picture cluster of each who object according to the picture comprised being arranged.
Picture obtains submodule 601d, and each picture be set to for a picture cluster in front n picture cluster being comprised is retrieved as multiple pictures comprising the face of same who object.
In sum, the device of the determination personage attribute information that disclosure embodiment provides, by corresponding to this user account, each the picture comprising face carries out recognition of face, obtain the picture cluster that each who object is corresponding, corresponding according to a user account, comprise the quantity determination numerical value n of the picture of face, the quantity order from big to small of the picture cluster of this each who object according to the picture comprised is arranged, and each picture that a picture cluster in front n picture cluster comprises is retrieved as these multiple comprise the picture of the face of same who object, when carrying out the identification of personage's attribute information in units of unique user account, ensure that the picture number comprising the face of same who object got enough meets the requirement of accuracy of identification, thus ensure the accuracy to the identification of personage's attribute information.
It should be noted is that, the device that above-described embodiment provides is when the method for personage's attribute information is determined in execution, only be illustrated with the division of each functional module above-mentioned, in practical application, can distribute according to actual needs and by above-mentioned functions and be completed by different functional modules, content structure by equipment is divided into different functional modules, to complete all or part of function described above.
About the device in above-described embodiment, wherein the concrete mode of modules executable operations has been described in detail in about the embodiment of the method, will not elaborate explanation herein.
The disclosure one exemplary embodiment additionally provides a kind of device determining personage's attribute information, can realize the method for the determination personage attribute information that disclosure embodiment provides.This device comprises: processor, and the storer of executable instruction for storage of processor.
Wherein, processor is configured to:
Obtain the picture that multiple comprise the face of same who object;
Respectively feature identification is carried out to the face of the who object comprised in plurality of pictures, determines each self-corresponding personage's attribute information of plurality of pictures; Personage's attribute information comprises age and/or sex;
In conjunction with each self-corresponding personage's attribute information of plurality of pictures, determine personage's attribute information of who object.
Alternatively, respectively feature identification is carried out to the face of the who object comprised in plurality of pictures, determines each self-corresponding personage's attribute information in plurality of pictures, comprising:
When personage's attribute information comprises the age, for each pictures in plurality of pictures, obtain the shooting time of picture;
Feature identification is carried out to the face of the who object comprised in picture, the age of who object when determining that picture is taken;
Age of who object and current time when being taken according to shooting time, picture, determine picture corresponding, age that character image is current;
By corresponding for picture, that character image is current age, be defined as the age that picture is corresponding.
Alternatively, in conjunction with each self-corresponding personage's attribute information of plurality of pictures, determine personage's attribute information of who object, comprising:
When personage's attribute information comprises the age, plurality of pictures each self-corresponding age is averaged, obtain the age of who object.
Alternatively, in conjunction with each self-corresponding personage's attribute information of plurality of pictures, determine personage's attribute information of who object, comprising:
When personage's attribute information comprises sex, determine a sex of large percentage shared in each self-corresponding sex of plurality of pictures;
A sex of large percentage shared in each self-corresponding sex of plurality of pictures is defined as the sex of who object.
Alternatively, the method also comprises:
Before a sex of large percentage shared in each self-corresponding sex of plurality of pictures is defined as the sex of who object, whether the ratio that a sex of the large percentage shared by detecting in each self-corresponding sex of plurality of pictures is corresponding is greater than predetermined threshold value;
If testing result is ratio be greater than predetermined threshold value, then perform a step sex of large percentage shared in each self-corresponding sex of plurality of pictures being defined as the sex of who object.
Alternatively, obtain the picture that multiple comprise the face of same who object, comprising:
According to user account corresponding, the quantity determination numerical value n of the picture that comprises face, n is the integer of >=1;
To user account corresponding, each the picture comprising face carry out recognition of face, obtain the picture cluster that each who object is corresponding, each pictures in a picture cluster all comprises the face of who object corresponding to picture cluster;
The quantity order from big to small of the picture cluster of each who object according to the picture comprised is arranged;
Each picture that a picture cluster in front n picture cluster comprises is retrieved as multiple pictures comprising the face of same who object.
Fig. 7 is the block diagram of a kind of device 700 according to an exemplary embodiment.Such as, device 700 may be provided in a server.With reference to Fig. 7, device 700 comprises processing components 722, and it comprises one or more processor further, and the memory resource representated by storer 732, such as, for storing the instruction that can be performed by processing element 722, application program.The application program stored in storer 732 can comprise each module corresponding to one group of instruction one or more.In addition, processing components 722 is configured to perform instruction, to perform the method shown in above-mentioned Fig. 2,3A or 4.
Device 700 can also comprise the power management that a power supply module 726 is configured to actuating unit 700, and a wired or wireless network interface 750 is configured to device 700 to be connected to network, and input and output (I/O) interface 758.Device 700 can operate the operating system based on being stored in storer 732, such as WindowsServerTM, MacOSXTM, UnixTM, LinuxTM, FreeBSDTM or similar.
Those skilled in the art, at consideration instructions and after putting into practice invention disclosed herein, will easily expect other embodiment of the present disclosure.The application is intended to contain any modification of the present disclosure, purposes or adaptations, and these modification, purposes or adaptations are followed general principle of the present disclosure and comprised the undocumented common practise in the art of the disclosure or conventional techniques means.Instructions and embodiment are only regarded as exemplary, and true scope of the present disclosure and spirit are pointed out by claim below.
Should be understood that, the disclosure is not limited to precision architecture described above and illustrated in the accompanying drawings, and can carry out various amendment and change not departing from its scope.The scope of the present disclosure is only limited by appended claim.
Claims (13)
1. determine a method for personage's attribute information, it is characterized in that, described method comprises:
Obtain the picture that multiple comprise the face of same who object;
Respectively feature identification is carried out to the face of the described who object comprised in described plurality of pictures, determines each self-corresponding personage's attribute information of described plurality of pictures; Described personage's attribute information comprises age and/or sex;
In conjunction with each self-corresponding personage's attribute information of described plurality of pictures, determine personage's attribute information of described who object.
2. method according to claim 1, is characterized in that, the face of the described described who object to comprising in described plurality of pictures carries out feature identification respectively, determines each self-corresponding personage's attribute information in described plurality of pictures, comprising:
When described personage's attribute information comprises the age, for each pictures in described plurality of pictures, obtain the shooting time of described picture;
Feature identification is carried out to the face of the described who object comprised in described picture, the age of described who object when determining that described picture is taken;
Age of described who object and current time when being taken according to described shooting time, described picture, determine described picture corresponding, age that described character image is current;
By corresponding for described picture, that described character image is current age, be defined as the age that described picture is corresponding.
3. method according to claim 1, is characterized in that, each self-corresponding personage's attribute information of plurality of pictures described in described combination, determines personage's attribute information of described who object, comprising:
When described personage's attribute information comprises the age, described plurality of pictures each self-corresponding age is averaged, obtains the age of described who object.
4. method according to claim 1, is characterized in that, each self-corresponding personage's attribute information of plurality of pictures described in described combination, determines personage's attribute information of described who object, comprising:
When described personage's attribute information comprises sex, determine a sex of large percentage shared in each self-corresponding sex of described plurality of pictures;
A sex of large percentage shared in each self-corresponding sex of described plurality of pictures is defined as the sex of described who object.
5. method according to claim 4, is characterized in that, described method also comprises:
Before a sex of large percentage shared in each self-corresponding sex of described plurality of pictures is defined as the sex of described who object, whether the ratio that a sex of the large percentage shared by detecting in each self-corresponding sex of described plurality of pictures is corresponding is greater than predetermined threshold value;
If testing result is described ratio be greater than described predetermined threshold value, then perform the step that a described sex by large percentage shared in each self-corresponding sex of described plurality of pictures is defined as the sex of described who object.
6. method according to claim 1, is characterized in that, described acquisition multiple comprise the picture of the face of same who object, comprising:
According to user account corresponding, the quantity determination numerical value n of the picture that comprises face, n is the integer of >=1;
To described user account corresponding, each the picture comprising face carry out recognition of face, obtain the picture cluster that each who object is corresponding, each pictures in a described picture cluster all comprises the face of who object corresponding to described picture cluster;
The quantity order from big to small of the picture cluster of each who object described according to the picture comprised is arranged;
Described in being retrieved as by each picture that a picture cluster in front n picture cluster comprises, multiple comprise the picture of the face of same who object.
7. determine a device for personage's attribute information, it is characterized in that, described device comprises:
Picture acquisition module, for obtaining the picture that multiple comprise the face of same who object;
First determination module, for carrying out feature identification respectively to the face of the described who object comprised in described plurality of pictures, determines each self-corresponding personage's attribute information of described plurality of pictures; Described personage's attribute information comprises age and/or sex;
Second determination module, in conjunction with each self-corresponding personage's attribute information of described plurality of pictures, determines personage's attribute information of described who object.
8. device according to claim 7, is characterized in that, described first determination module, comprising:
Time obtains submodule, for when described personage's attribute information comprises the age, for each pictures in described plurality of pictures, obtains the shooting time of described picture;
First age determined submodule, for carrying out feature identification to the face of the described who object comprised in described picture, and the age of described who object when determining that described picture is taken;
Second age determined submodule, age of described who object and current time during for being taken according to described shooting time, described picture, determine described picture corresponding, age that described character image is current;
3rd age determined submodule, for by corresponding for described picture, that described character image is current age, was defined as the age that described picture is corresponding.
9. device according to claim 7, is characterized in that, described second determination module, comprising:
Mean value calculation submodule, for when described personage's attribute information comprises the age, averages to described plurality of pictures each self-corresponding age, obtains the age of described who object.
10. device according to claim 7, is characterized in that, described second determination module, comprising:
First sex determination submodule, for when described personage's attribute information comprises sex, determines a sex of large percentage shared in each self-corresponding sex of described plurality of pictures;
Second sex determination submodule, for being defined as the sex of described who object by a sex of large percentage shared in each self-corresponding sex of described plurality of pictures.
11. devices according to claim 10, is characterized in that, described knot second determination module, also comprises:
Detection sub-module, before a sex of large percentage shared in each self-corresponding sex of described plurality of pictures being defined as the sex of described who object at described second sex determination submodule, whether the ratio that a sex of the large percentage shared by detecting in each self-corresponding sex of described plurality of pictures is corresponding is greater than predetermined threshold value;
Described second sex determination submodule, for being that described ratio is greater than described predetermined threshold value in the testing result of described detection sub-module, perform the step that a described sex by large percentage shared in each self-corresponding sex of described plurality of pictures is defined as the sex of described who object.
12. devices according to claim 7, is characterized in that, described picture acquisition module, comprising:
Numerical value determination submodule, for according to user account corresponding, the quantity determination numerical value n of the picture that comprises face, n is the integer of >=1;
Cluster submodule, for to described user account corresponding, each the picture comprising face carry out recognition of face, obtain the picture cluster that each who object is corresponding, each pictures in a described picture cluster all comprises the face of who object corresponding to described picture cluster;
Sorting sub-module, for arranging the quantity order from big to small of the picture cluster of each who object described according to the picture comprised;
Picture obtains submodule, and described in each picture for a picture cluster in front n picture cluster being comprised is retrieved as, multiple comprise the picture of the face of same who object.
13. 1 kinds of devices determining personage's attribute information, it is characterized in that, described device comprises:
Processor;
For storing the storer of the executable instruction of described processor;
Wherein, described processor is configured to:
Obtain the picture that multiple comprise the face of who object;
Respectively feature identification is carried out to the face of the described who object comprised in described plurality of pictures, determines each self-corresponding personage's attribute information of described plurality of pictures; Described personage's attribute information comprises age and/or sex;
In conjunction with each self-corresponding personage's attribute information of described plurality of pictures, determine personage's attribute information of described who object.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510857441.8A CN105488470A (en) | 2015-11-30 | 2015-11-30 | Method and apparatus for determining character attribute information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510857441.8A CN105488470A (en) | 2015-11-30 | 2015-11-30 | Method and apparatus for determining character attribute information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105488470A true CN105488470A (en) | 2016-04-13 |
Family
ID=55675442
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510857441.8A Pending CN105488470A (en) | 2015-11-30 | 2015-11-30 | Method and apparatus for determining character attribute information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105488470A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108021882A (en) * | 2017-12-01 | 2018-05-11 | 宁波高新区锦众信息科技有限公司 | A kind of face identification system of the robot based on the period |
| CN109463897A (en) * | 2018-12-28 | 2019-03-15 | 蒋梦兰 | Brushing mode replacement system |
| CN109598176A (en) * | 2017-09-30 | 2019-04-09 | 佳能株式会社 | Identification device and recognition methods |
| CN110084267A (en) * | 2019-03-12 | 2019-08-02 | 北京旷视科技有限公司 | Portrait clustering method, device, electronic equipment and readable storage medium storing program for executing |
| CN110473171A (en) * | 2019-07-18 | 2019-11-19 | 上海联影智能医疗科技有限公司 | Brain age detection method, computer equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102592131A (en) * | 2011-01-06 | 2012-07-18 | 上海银晨智能识别科技有限公司 | Device for comparing a plurality of photos of the same user |
| CN103886284A (en) * | 2014-03-03 | 2014-06-25 | 小米科技有限责任公司 | Character attribute information identification method and device and electronic device |
| CN105069083A (en) * | 2015-07-31 | 2015-11-18 | 小米科技有限责任公司 | Determination method and device of associated user |
| CN105095873A (en) * | 2015-07-31 | 2015-11-25 | 小米科技有限责任公司 | Picture sharing method and apparatus |
-
2015
- 2015-11-30 CN CN201510857441.8A patent/CN105488470A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102592131A (en) * | 2011-01-06 | 2012-07-18 | 上海银晨智能识别科技有限公司 | Device for comparing a plurality of photos of the same user |
| CN103886284A (en) * | 2014-03-03 | 2014-06-25 | 小米科技有限责任公司 | Character attribute information identification method and device and electronic device |
| CN105069083A (en) * | 2015-07-31 | 2015-11-18 | 小米科技有限责任公司 | Determination method and device of associated user |
| CN105095873A (en) * | 2015-07-31 | 2015-11-25 | 小米科技有限责任公司 | Picture sharing method and apparatus |
Non-Patent Citations (2)
| Title |
|---|
| 吴萌: "多尺度表达和正则化方法在图像识别中的研究与应用", 《中国博士学位论文全文数据库》 * |
| 李杨: "基于多任务学习的人脸图像识别", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109598176A (en) * | 2017-09-30 | 2019-04-09 | 佳能株式会社 | Identification device and recognition methods |
| CN108021882A (en) * | 2017-12-01 | 2018-05-11 | 宁波高新区锦众信息科技有限公司 | A kind of face identification system of the robot based on the period |
| CN108021882B (en) * | 2017-12-01 | 2021-12-21 | 宁波高新区锦众信息科技有限公司 | Face recognition system of robot based on time interval |
| CN109463897A (en) * | 2018-12-28 | 2019-03-15 | 蒋梦兰 | Brushing mode replacement system |
| CN110084267A (en) * | 2019-03-12 | 2019-08-02 | 北京旷视科技有限公司 | Portrait clustering method, device, electronic equipment and readable storage medium storing program for executing |
| CN110473171A (en) * | 2019-07-18 | 2019-11-19 | 上海联影智能医疗科技有限公司 | Brain age detection method, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105488470A (en) | Method and apparatus for determining character attribute information | |
| CN105005980B (en) | Image processing method and device | |
| US9607394B2 (en) | Information processing method and electronic device | |
| CN110516678B (en) | Image processing method and device | |
| CN109784304B (en) | Method and apparatus for labeling dental images | |
| CN113704531A (en) | Image processing method, image processing device, electronic equipment and computer readable storage medium | |
| CN103765421A (en) | Content control method, content control apparatus, and program | |
| CN112732975B (en) | Object tracking method and device, electronic equipment and system | |
| US11210563B2 (en) | Method and apparatus for processing image | |
| CN111199540A (en) | Image quality evaluation method, image quality evaluation device, electronic device, and storage medium | |
| CN112559884A (en) | Method and device for hooking panorama and interest point, electronic equipment and storage medium | |
| CN109919220B (en) | Method and apparatus for generating feature vectors of video | |
| CN108921138B (en) | Method and apparatus for generating information | |
| CN114266937A (en) | Model training method, image processing method, device, equipment and storage medium | |
| US11429660B2 (en) | Photo processing method, device and computer equipment | |
| CN111049735B (en) | Group head portrait display method, device, equipment and storage medium | |
| CN117610524A (en) | Form generation method, form generation device, electronic equipment and computer readable storage medium | |
| CN110413869B (en) | Method and device for pushing information | |
| CN114881227B (en) | Model compression method, image processing device and electronic equipment | |
| CN113610856B (en) | Method and device for training image segmentation model and image segmentation | |
| CN113938751B (en) | Video transition type determining method, device and storage medium | |
| CN105991400B (en) | Group searching method and device | |
| CN112598074A (en) | Image processing method and device, computer readable storage medium and electronic device | |
| CN111694931A (en) | Element acquisition method and device | |
| CN113420176B (en) | Question searching method, question frame drawing device, question searching equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into 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: 20160413 |