CN103996032A - Face angle determining method based on cranium image coincidence theory - Google Patents

Face angle determining method based on cranium image coincidence theory Download PDF

Info

Publication number
CN103996032A
CN103996032A CN201410227406.3A CN201410227406A CN103996032A CN 103996032 A CN103996032 A CN 103996032A CN 201410227406 A CN201410227406 A CN 201410227406A CN 103996032 A CN103996032 A CN 103996032A
Authority
CN
China
Prior art keywords
face
point
distance
deflection
nasion
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
Application number
CN201410227406.3A
Other languages
Chinese (zh)
Inventor
詹东晖
文正
卢磊
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xiamen Reconova Information Technology Co Ltd
Original Assignee
Xiamen Reconova Information Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Xiamen Reconova Information Technology Co Ltd filed Critical Xiamen Reconova Information Technology Co Ltd
Priority to CN201410227406.3A priority Critical patent/CN103996032A/en
Publication of CN103996032A publication Critical patent/CN103996032A/en
Pending legal-status Critical Current

Links

Landscapes

  • Image Processing (AREA)

Abstract

The invention discloses a face angle determining method based on a cranium image coincidence theory. The face angle determining method is mainly characterized in that face deflection angles are calculated through two-dimensional facial feature points and include the face horizontal deflection angle and the face pitching deflection angle. By means of the face angle determining method, the face angles are determined through a face deflection index according to the cranium image coincidence theory, and are accurately estimated through the two-dimensional facial feature points; in addition, due to the fact that the face deflection index is used for calculation, limitation of the size of a picture is avoided, and values of the deflection angles can be rapidly found through a convertible regression equation and a convertible curve.

Description

A kind of face angle decision method based on photographic superimposition theory
Technical field
The present invention relates to a kind of face angle decision method based on photographic superimposition theory.
Background technology
The method of both at home and abroad human face posture being estimated at present has three classes substantially: the method based on model, the method based on face outward appearance and the method based on classification.
Method based on model is to utilize geometric model and the unique point of face, between model and image, sets up corresponding relation, then realizes the estimation of human face posture by set or other method.First carry out the detection of unique point, generally adopt eyes, nose and mouth; Then face is set up to three-dimensional model, three-dimensional feature is put to continuous projection and approach the unique point on image; Adopt simple video camera to photograph due to general in reality, be not easy to set up the three-dimensional model of face.
Method based on face outward appearance is that the content of hypothesis human face posture and image exists mutual relation, such as gray scale, gradient etc., then, by a large amount of sample of training, finds out this corresponding relation, thereby carries out the estimation of attitude.Due to a large amount of training sample image of needs, different attitude, illumination all can exert an influence to the confirmation of relation, more complicated.
Method based on classification is to adopt the positive negative sample of training, obtains a statistical value according to sample, and utilizes regression algorithm to calculate regression equation this value, thereby the calculating of attitude is converted to the calculating of regression equation, obtains general angle.Although the method based on classification can find the change direction of human face posture easily, estimate human face posture or some difficulty by this parameter.
Summary of the invention
The object of the present invention is to provide a kind of face angle decision method based on photographic superimposition theory, can utilize two-dimentional face feature point more accurately to estimate face deflection angle.
A face angle decision method based on photographic superimposition theory, comprises the steps:
Step 1, the two-dimentional face feature point of facial image is carried out to coordinate setting, obtain the coordinate figure of left outside eye Le, right outside eye Re, nasion o, glabella point G, subnasal point Nd and gnathion Cd, calculate obtain distance R e-o, the glabella point G of left outside eye Le to the distance L e-o of nasion o, right outside Re to nasion o to the distance G-Nd of subnasal point Nd and subnasal point Nd to gnathion Cd apart from Nd-Cd;
If the left outside eye of step 2 Le equals the distance R e-o of right outside eye Re to nasion o to the distance L e-o of nasion o, judge that face X deflection angle is as 0;
If left outside eye Le is not equal to the distance R e-o of right outside eye Re to nasion o to the distance L e-o of nasion o, horizontal deflection index X=Le-o/Re-o, by the regression equation Y=91.33-89.09X of face horizontal deflection index X and X deflection angle Y, calculate the value of face X deflection angle Y, when calculate result n>0 time, the n ° that shown deflection left, when result of calculation n<0, the n ° that shown deflection to the right;
?if step 3 glabella point G to the distance G-Nd of subnasal point Nd equal subnasal point Nd to gnathion Cd apart from Nd-Cd, judge that face pitching deflection angle is 0; If glabella point G to the distance G-Nd of subnasal point Nd be not equal to subnasal point Nd to gnathion Cd apart from Nd-Cd, pitching index X=G-Nd/Nd-Cd, by the regression equation Y=94.54-98.70X of pitching index X and luffing angle Y, calculate the value of face pitching deflection angle Y, when calculate result n>0 time, show to have upward deflected n °, when result of calculation n<0, show to have deflected down n °.
Portrait is all consistent (pitching is also identical) with the horizontal deflection index of skull picture on various deflection angles, the present invention carries out the judgement of face angle according to photographic superimposition theory, can utilize two-dimentional face feature point more accurately to estimate face deflection angle, and because using face tilt index, the present invention calculates, be not subject to the restriction of photo size, by can, for the regression equation and the curve that convert, checking in rapidly deflection angle value.
Brief description of the drawings
Fig. 1 is the graph of a relation of the flat tilt index X of inventor's water for washing the face and X deflection angle Y;
Fig. 2 is the graph of a relation of face pitching index X of the present invention and luffing angle Y.
Below in conjunction with the drawings and specific embodiments, the present invention is further described.
Embodiment
A kind of face angle decision method based on photographic superimposition theory of the present invention, is mainly to utilize two-dimentional face feature point to calculate face deflection angle, is divided into face X deflection angle and face pitching deflection angle.
1, calculate and obtain face X deflection angle according to face horizontal deflection index:
In general, face is symmetrical, so left outside eye Le should equate to the distance R e-o of nasion o to distance L e-o and the right outside eye Re of nasion o, for left outside eye Le and right outside eye Re do a connecting line, when face is during just to camera lens, its X deflection angle is 0 degree, at this moment the distance R e-o of left outside eye Le to the distance L e-o of nasion o and right outside Re to nasion o is equal in length, when X deflection angle is not equal to 0 while spending, according to the rule of perspective projection, can be greater than from the length after the side line section imaging close to camera lens that (side away from camera lens is called deflection side from the side away from camera lens, a side close to camera lens is called deflection offside), horizontal deflection index X=Le-o/Re-o so,
As shown in Figure 1, obtain the inverse ratio linear relation of face horizontal deflection index X and X deflection angle Y through statistical test, its regression equation is: Y=91.33-89.09X;
When calculate result n>0 time, the n ° that shown deflection left; When result of calculation n<0, the n ° that shown deflection to the right;
2, calculate and obtain face luffing angle according to face pitching index:
If glabella point G, subnasal point Nd, gnathion Cd, if glabella point G to the distance G-Nd of subnasal point Nd equal subnasal point Nd to gnathion Cd apart from Nd-Cd, judge that face pitching deflection angle is 0, if glabella point G to the distance G-Nd of subnasal point Nd be not equal to subnasal point Nd to gnathion Cd apart from Nd-Cd, pitching index X=G-Nd/Nd-Cd;
When calculate result n>0 time, show to have upward deflected n °; When result of calculation n<0, show to have deflected down n °
As shown in Figure 2, obtain the inverse ratio linear relation of pitching index X and luffing angle Y through statistical test, its regression equation is: Y=94.54-98.70X.
A kind of face angle decision method based on photographic superimposition theory of the present invention, specifically comprises the steps:
Step 1, the two-dimentional face feature point of facial image is carried out to coordinate setting, obtain the coordinate figure of left outside eye Le, right outside eye Re, nasion o, glabella point G, subnasal point Nd and gnathion Cd, calculate obtain distance R e-o, the glabella point G of left outside eye Le to the distance L e-o of nasion o, right outside Re to nasion o to the distance G-Nd of subnasal point Nd and subnasal point Nd to gnathion Cd apart from Nd-Cd;
If the left outside eye of step 2 Le equals the distance R e-o of right outside eye Re to nasion o to the distance L e-o of nasion o, judge that face X deflection angle is as 0;
If left outside eye Le is not equal to the distance R e-o of right outside eye Re to nasion o to the distance L e-o of nasion o, horizontal deflection index X=Le-o/Re-o, by the regression equation Y=91.33-89.09X of face horizontal deflection index X and X deflection angle Y, calculate the value of face X deflection angle Y; When calculate result n>0 time, the n ° that shown deflection left; When result of calculation n<0, the n ° that shown deflection to the right;
?if step 3 glabella point G to the distance G-Nd of subnasal point Nd equal subnasal point Nd to gnathion Cd apart from Nd-Cd, judge that face pitching deflection angle is 0; If glabella point G to the distance G-Nd of subnasal point Nd be not equal to subnasal point Nd to gnathion Cd apart from Nd-Cd, pitching index X=G-Nd/Nd-Cd, by the regression equation Y=94.54-98.70X of pitching index X and luffing angle Y, calculate the value of face pitching deflection angle Y, when calculate result n>0 time, show to have upward deflected n °; When result of calculation n<0, show to have deflected down n °.
The above, it is only preferred embodiment of the present invention, not technical scope of the present invention is imposed any restrictions, therefore any trickle amendment, equivalent variations and modification that every foundation technical spirit of the present invention is done above embodiment all still belong in the scope of technical solution of the present invention.

Claims (1)

1. the face angle decision method based on photographic superimposition theory, is characterized in that comprising the steps:
Step 1, the two-dimentional face feature point of facial image is carried out to coordinate setting, obtain the coordinate figure of left outside eye Le, right outside eye Re, nasion o, glabella point G, subnasal point Nd and gnathion Cd, calculate obtain distance R e-o, the glabella point G of left outside eye Le to the distance L e-o of nasion o, right outside Re to nasion o to the distance G-Nd of subnasal point Nd and subnasal point Nd to gnathion Cd apart from Nd-Cd;
If the left outside eye of step 2 Le equals the distance R e-o of right outside eye Re to nasion o to the distance L e-o of nasion o, judge that face X deflection angle is as 0;
If left outside eye Le is not equal to the distance R e-o of right outside eye Re to nasion o to the distance L e-o of nasion o, horizontal deflection index X=Le-o/Re-o, by the regression equation Y=91.33-89.09X of face horizontal deflection index X and X deflection angle Y, calculate the value of face X deflection angle Y, when calculate result n>0 time, the n ° that shown deflection left, when result of calculation n<0, the n ° that shown deflection to the right;
?if step 3 glabella point G to the distance G-Nd of subnasal point Nd equal subnasal point Nd to gnathion Cd apart from Nd-Cd, judge that face pitching deflection angle is 0; If glabella point G to the distance G-Nd of subnasal point Nd be not equal to subnasal point Nd to gnathion Cd apart from Nd-Cd, pitching index X=G-Nd/Nd-Cd, by the regression equation Y=94.54-98.70X of pitching index X and luffing angle Y, calculate the value of face pitching deflection angle Y, when calculate result n>0 time, show to have upward deflected n °, when result of calculation n<0, show to have deflected down n °.
CN201410227406.3A 2014-05-27 2014-05-27 Face angle determining method based on cranium image coincidence theory Pending CN103996032A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410227406.3A CN103996032A (en) 2014-05-27 2014-05-27 Face angle determining method based on cranium image coincidence theory

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410227406.3A CN103996032A (en) 2014-05-27 2014-05-27 Face angle determining method based on cranium image coincidence theory

Publications (1)

Publication Number Publication Date
CN103996032A true CN103996032A (en) 2014-08-20

Family

ID=51310193

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410227406.3A Pending CN103996032A (en) 2014-05-27 2014-05-27 Face angle determining method based on cranium image coincidence theory

Country Status (1)

Country Link
CN (1) CN103996032A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105389570A (en) * 2015-11-19 2016-03-09 吴建忠 Face angle determination method and system
CN109522775A (en) * 2017-09-19 2019-03-26 杭州海康威视数字技术股份有限公司 Face character detection method, device and electronic equipment
CN115331292A (en) * 2022-08-17 2022-11-11 武汉元紫东科技有限公司 Facial image-based emotion recognition method and device and computer storage medium

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103793693A (en) * 2014-02-08 2014-05-14 厦门美图网科技有限公司 Method for detecting face turning and facial form optimizing method with method for detecting face turning

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103793693A (en) * 2014-02-08 2014-05-14 厦门美图网科技有限公司 Method for detecting face turning and facial form optimizing method with method for detecting face turning

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
刘晓宁: "计算机辅助颅像重合技术的研究与实现", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105389570A (en) * 2015-11-19 2016-03-09 吴建忠 Face angle determination method and system
CN109522775A (en) * 2017-09-19 2019-03-26 杭州海康威视数字技术股份有限公司 Face character detection method, device and electronic equipment
CN109522775B (en) * 2017-09-19 2021-07-20 杭州海康威视数字技术股份有限公司 Face attribute detection method and device and electronic equipment
CN115331292A (en) * 2022-08-17 2022-11-11 武汉元紫东科技有限公司 Facial image-based emotion recognition method and device and computer storage medium

Similar Documents

Publication Publication Date Title
JP6742405B2 (en) Head-mounted display with facial expression detection function
EP3323249B1 (en) Three dimensional content generating apparatus and three dimensional content generating method thereof
CN105894574B (en) A kind of binocular three-dimensional reconstruction method
CN104173054B (en) Measuring method and measuring device for height of human body based on binocular vision technique
CN109446892B (en) Human eye attention positioning method and system based on deep neural network
KR101977638B1 (en) Method for correcting user’s gaze direction in image, machine-readable storage medium and communication terminal
KR101874494B1 (en) Apparatus and method for calculating 3 dimensional position of feature points
TWI553565B (en) Utilizing two-dimensional image to estimate its three-dimensional face angle method, and its database establishment of face replacement and face image replacement method
US11534063B2 (en) Interpupillary distance measuring method, wearable ophthalmic device and storage medium
CN111414780A (en) Sitting posture real-time intelligent distinguishing method, system, equipment and storage medium
CA2794659A1 (en) Apparatus and method for iris recognition using multiple iris templates
CN109035309A (en) Pose method for registering between binocular camera and laser radar based on stereoscopic vision
CN106570899B (en) Target object detection method and device
GB201308377D0 (en) A system for mixing or composting in real-time, computer generated 3D objects and a video feed from a film camera
CN106033614B (en) A kind of mobile camera motion object detection method under strong parallax
CN106778660B (en) A kind of human face posture bearing calibration and device
CN104317385A (en) Gesture identification method and system
JP2012069133A (en) Solidity authenticating method, solidity authenticating apparatus, and solidity authenticating program
CN105740778A (en) Improved three-dimensional human face in-vivo detection method and device thereof
CN108748149A (en) Based on deep learning without calibration mechanical arm grasping means under a kind of complex environment
US20190156511A1 (en) Region of interest image generating device
CN103996032A (en) Face angle determining method based on cranium image coincidence theory
CN103900473A (en) Intelligent mobile device six-degree-of-freedom fused pose estimation method based on camera and gravity inductor
CN104123562A (en) Human body face expression identification method and device based on binocular vision
CN109359537A (en) Human face posture angle detecting method neural network based and system

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
WD01 Invention patent application deemed withdrawn after publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20140820