CN109102533A - A kind of characteristic point positioning method based on mixed reality - Google Patents
A kind of characteristic point positioning method based on mixed reality Download PDFInfo
- Publication number
- CN109102533A CN109102533A CN201810630885.1A CN201810630885A CN109102533A CN 109102533 A CN109102533 A CN 109102533A CN 201810630885 A CN201810630885 A CN 201810630885A CN 109102533 A CN109102533 A CN 109102533A
- Authority
- CN
- China
- Prior art keywords
- mixed reality
- model
- point cloud
- characteristic point
- human body
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
- G06T7/344—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
- G06T2207/30201—Face
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Processing Or Creating Images (AREA)
Abstract
The present invention relates to a kind of characteristic point positioning methods based on mixed reality, method includes the following steps: (1) records point cloud data by acquisition human body feature point;(2) point cloud data of feature is generated into two dimensional code;(3) two dimensional code is identified by the camera that mixed reality equipment carries, obtains feature point cloud model;(4) by carrying out aspect ratio pair with the model in mixed reality equipment, it is registrated simultaneously computation model error, differentiates the similarity between human body and model.Human body feature point is constructed point cloud data by the present invention, then the identifiable two dimensional code of mixed reality equipment is generated, then it is compared with the model prestored in mixed reality equipment, reach distinguish the model whether be this human body information data, solve the problems, such as that mixed reality model is matched with patient status's information.
Description
Technical field
The invention belongs to image recognitions and field of medical image processing, are related to a kind of positioning feature point based on mixed reality
Method.
Background technique
Mixed reality equipment is the first holographic computer equipment not limited by cable of Microsoft, can allow user and digital content
Interaction, and interacted with the hologram in true environment around.It may be implemented to advise preoperative simulation by mixed reality at present
It draws and is interacted with across space remote operation, and then implement precisely operation, greatly reduce the risk of operation.When same set of mixed reality
For device service when multiple sufferers, which needs the problem of should matching to mixed reality model with patient status's information,
If cannot correctly match, result is difficult to expect by bringing.So the present invention passes through a kind of feature based on mixed reality
Independent positioning method, to solve the problems, such as this.
Summary of the invention
The object of the present invention is to provide a kind of characteristic point positioning method based on mixed reality, solve mixed reality model with
The problem of patient status's information matches.
The present invention is achieved through the following technical solutions: a kind of characteristic point positioning method based on mixed reality, this method
The following steps are included:
(1) by acquisition human body feature point, point cloud data is recorded;
(2) point cloud data of feature is generated into two dimensional code;
(3) two dimensional code is identified by the camera that mixed reality equipment carries, obtains feature point cloud model;
(4) by carrying out aspect ratio pair with the model in mixed reality equipment, it is registrated simultaneously computation model error, differentiates human body
Similarity between model.
Further, the construction method of point cloud data is to establish cartesian coordinate system in step (1), drafts arbitrary origin,
The coordinate information of characteristic point is recorded, point cloud data is formed.
Further, the aspect ratio pair in step (4), the comparison picture that mixed reality equipment uploads are 3 dimensions, and characteristic point is raw
At contrast images be similarly 3 dimensions, calculate the error of two groups of 3 d image models, by the method for point cloud registering for judging
Whether human body information coincide with the model in mixed reality equipment.
Further, the method for the point cloud registering is ICP algorithm.
Further, the mixed reality equipment is Microsoft Hololens glasses.
Good effect by adopting the above technical scheme: human body feature point is constructed point cloud data by the present invention, is then generated mixed
The identifiable two dimensional code of real world devices is closed, is then compared with the model prestored in mixed reality equipment, reaches and distinguishes the mould
Type whether be this human body information data, solve the problems, such as that mixed reality model is matched with patient status's information.
Detailed description of the invention
Fig. 1 is for facial image features point distribution schematic diagram;
Fig. 2 is to save the three dimensional face model for being patient in mixing apparatus;
Fig. 3 is the characteristic point positioning method schematic diagram of the invention based on mixed reality:
Fig. 4 is ICP registration result schematic diagram;
Fig. 5 is the error calculation schematic diagram of registration.
Specific embodiment
The following further describes the technical solution of the present invention with reference to the accompanying drawing, but should not be construed as to limit of the invention
System:
A kind of characteristic point positioning method based on mixed reality, method includes the following steps:
(1) by acquisition human body feature point, point cloud data is recorded;
(2) point cloud data of feature is generated into two dimensional code;
(3) two dimensional code is identified by the camera that mixed reality equipment carries, obtains feature point cloud model;
(4) by carrying out aspect ratio pair with the model in mixed reality equipment, it is registrated simultaneously computation model error, differentiates human body
Similarity between model.
Further, the construction method of point cloud data is to establish cartesian coordinate system in step (1), drafts arbitrary origin,
The coordinate information of characteristic point is recorded, point cloud data is formed.
Further, the aspect ratio pair in step (4), the comparison picture that mixed reality equipment uploads are 3 dimensions, and characteristic point is raw
At contrast images be similarly 3 dimensions, calculate the error of two groups of 3 d image models, by the method for point cloud registering for judging
Whether human body information coincide with the model in mixed reality equipment.
Further, the method for the point cloud registering is ICP algorithm.
Further, the mixed reality equipment is Microsoft Hololens glasses.
Embodiment 1
A kind of extraction facial image features point data method includes:
Step 1: wherein characteristic point is distributed using 90 positioning feature point faces are as follows: 18 points mark mouth, 14 points
Label lower jaw, 12 points label eyes, 6 points label eyebrows, 4 points label cheek and cheek, 10 points label noses, 4
Point label posterior neck, 10 points mark ear, and 12 points mark hair.As shown in Figure 1.
Step 2: establishing cartesian coordinate system, arbitrary origin is drafted, records the coordinate information of characteristic point, building point cloud number
According to.The point cloud data is the coordinate array for possessing 90 rows 3 column, and 3 column distributions correspond to x, y, z coordinate value, and every row indicates different spies
Sign point.
It is stored Step 3: point cloud data is converted into two dimensional code, facilitates mixed reality equipment scanning recognition.
Embodiment 2
A kind of characteristic point positioning method based on mixed reality, comprising:
Step 1: doctor has the two dimensional code of patient characteristics by wearing mixed reality equipment, scanning strip, characteristic point cloud is obtained
Model, process are as shown in Figure 3.
Step 2: patient characteristics point cloud is named as p, the point cloud of model is named as q.A cloud is calculated by following formula
Center of gravity:
Covariance matrix is constructed using the point cloud center of gravity acquired:
Step 3: next entering ICP algorithm carries out Model registration, the function of ICP is can to match two number differences
Data acquisition system, then one 4 × 4 symmetrical matrix will be constructed using covariance matrix by realizing using ICP:
Wherein, it can be used to calculate rotation parameter by the maximal eigenvector of this symmetrical matrix, and then find out entire rigid
R required for body converts and T (R is best rotating vector, and T is best motion vector).The mode for obtaining selection translation can
To be registrated two images, and then achieve the effect that calculate error, registration result is as shown in Figure 4.
Step 4: error calculation mode will be illustrated next, as shown in figure 5, MODEL is p point cloud, DATA is q point cloud, mid
For the position of the intermediate point of DATA point cloud, iclosest table records the number by the matched point of point in each DATA cloud
(number in MODEL point cloud).
Step 5: being guaranteed by calculating MODLE point to the Euclidean distance of intermediate position points cloud registration point in the cloud
A section in the Euclidean distance of arbitrary point to DATA point cloud be both less than the distance that this point is generated.Similarly calculate
Each matched minimum Eustachian distance of point institute of mid ± n, and add up to these distances, obtained result is as mistake
Difference returns.
Human body feature point is constructed point cloud data by the present invention, then generates the identifiable two dimensional code of mixed reality equipment, so
Be compared afterwards with the model prestored in mixed reality equipment, reach distinguish the model whether be this human body information data,
Solve the problems, such as that mixed reality model is matched with patient status's information.
Claims (5)
1. a kind of characteristic point positioning method based on mixed reality, it is characterised in that: method includes the following steps:
(1) by acquisition human body feature point, point cloud data is recorded;
(2) point cloud data of feature is generated into two dimensional code;
(3) two dimensional code is identified by the camera that mixed reality equipment carries, obtains feature point cloud model;
(4) by carrying out aspect ratio pair with the model in mixed reality equipment, it is registrated simultaneously computation model error, differentiates human body and mould
Similarity between type.
2. the characteristic point positioning method according to claim 1 based on mixed reality, it is characterised in that: step (1) midpoint
The construction method of cloud data is to establish cartesian coordinate system, drafts arbitrary origin, records the coordinate information of characteristic point, forms point cloud
Data.
3. the characteristic point positioning method according to claim 1 based on mixed reality, it is characterised in that: in step (4)
Aspect ratio pair, the comparison picture that mixed reality equipment uploads are 3 dimensions, and the contrast images that characteristic point generates similarly are 3 dimensions, are passed through
The method of point cloud registering calculates the error of two groups of 3 d image models, for judging the mould in human body information and mixed reality equipment
Whether type coincide.
4. the characteristic point positioning method according to claim 3 based on mixed reality, it is characterised in that: the point cloud is matched
Quasi- method is ICP algorithm.
5. the characteristic point positioning method based on mixed reality described in -4 any one claims according to claim 1, special
Sign is: the mixed reality equipment is Microsoft Hololens glasses.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810630885.1A CN109102533A (en) | 2018-06-19 | 2018-06-19 | A kind of characteristic point positioning method based on mixed reality |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810630885.1A CN109102533A (en) | 2018-06-19 | 2018-06-19 | A kind of characteristic point positioning method based on mixed reality |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109102533A true CN109102533A (en) | 2018-12-28 |
Family
ID=64796949
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810630885.1A Pending CN109102533A (en) | 2018-06-19 | 2018-06-19 | A kind of characteristic point positioning method based on mixed reality |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109102533A (en) |
Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103400119A (en) * | 2013-07-31 | 2013-11-20 | 南京融图创斯信息科技有限公司 | Face recognition technology-based mixed reality spectacle interactive display method |
CN103854270A (en) * | 2012-11-28 | 2014-06-11 | 广州医学院第一附属医院 | CT and MR inter-machine three dimensional image fusion registration method and system |
CN103971079A (en) * | 2013-01-28 | 2014-08-06 | 腾讯科技(深圳)有限公司 | Augmented reality implementation method and device of two-dimensional code |
CN104376334A (en) * | 2014-11-12 | 2015-02-25 | 上海交通大学 | Pedestrian comparison method based on multi-scale feature fusion |
CN105188516A (en) * | 2013-03-11 | 2015-12-23 | 奇跃公司 | System and method for augmented and virtual reality |
CN105469042A (en) * | 2015-11-20 | 2016-04-06 | 天津汉光祥云信息科技有限公司 | Improved face image comparison method |
CN106248035A (en) * | 2016-08-19 | 2016-12-21 | 苏州大学 | The method and system that a kind of surface profile based on point cloud model accurately detects |
CN106973569A (en) * | 2014-05-13 | 2017-07-21 | Pcp虚拟现实股份有限公司 | Generation and the playback multimedia mthods, systems and devices of virtual reality |
CN107223269A (en) * | 2016-12-29 | 2017-09-29 | 深圳前海达闼云端智能科技有限公司 | Three-dimensional scene positioning method and device |
CN107678537A (en) * | 2017-09-04 | 2018-02-09 | 全球能源互联网研究院有限公司 | Assembly manipulation, the method and apparatus of simulation assembling are identified in augmented reality environment |
CN107798702A (en) * | 2016-08-30 | 2018-03-13 | 成都理想境界科技有限公司 | A kind of realtime graphic stacking method and device for augmented reality |
US20180103213A1 (en) * | 2016-10-06 | 2018-04-12 | Fyusion, Inc. | Live style transfer on a mobile device |
-
2018
- 2018-06-19 CN CN201810630885.1A patent/CN109102533A/en active Pending
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103854270A (en) * | 2012-11-28 | 2014-06-11 | 广州医学院第一附属医院 | CT and MR inter-machine three dimensional image fusion registration method and system |
CN103971079A (en) * | 2013-01-28 | 2014-08-06 | 腾讯科技(深圳)有限公司 | Augmented reality implementation method and device of two-dimensional code |
CN105188516A (en) * | 2013-03-11 | 2015-12-23 | 奇跃公司 | System and method for augmented and virtual reality |
CN103400119A (en) * | 2013-07-31 | 2013-11-20 | 南京融图创斯信息科技有限公司 | Face recognition technology-based mixed reality spectacle interactive display method |
CN106973569A (en) * | 2014-05-13 | 2017-07-21 | Pcp虚拟现实股份有限公司 | Generation and the playback multimedia mthods, systems and devices of virtual reality |
CN104376334A (en) * | 2014-11-12 | 2015-02-25 | 上海交通大学 | Pedestrian comparison method based on multi-scale feature fusion |
CN105469042A (en) * | 2015-11-20 | 2016-04-06 | 天津汉光祥云信息科技有限公司 | Improved face image comparison method |
CN106248035A (en) * | 2016-08-19 | 2016-12-21 | 苏州大学 | The method and system that a kind of surface profile based on point cloud model accurately detects |
CN107798702A (en) * | 2016-08-30 | 2018-03-13 | 成都理想境界科技有限公司 | A kind of realtime graphic stacking method and device for augmented reality |
US20180103213A1 (en) * | 2016-10-06 | 2018-04-12 | Fyusion, Inc. | Live style transfer on a mobile device |
CN107223269A (en) * | 2016-12-29 | 2017-09-29 | 深圳前海达闼云端智能科技有限公司 | Three-dimensional scene positioning method and device |
CN107678537A (en) * | 2017-09-04 | 2018-02-09 | 全球能源互联网研究院有限公司 | Assembly manipulation, the method and apparatus of simulation assembling are identified in augmented reality environment |
Non-Patent Citations (1)
Title |
---|
赵春雨: "基于复杂点云的三维重建", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20200192464A1 (en) | Gesture recognition using multi-sensory data | |
Gupta et al. | Texas 3D face recognition database | |
JP4950787B2 (en) | Image processing apparatus and method | |
CN110503703A (en) | Method and apparatus for generating image | |
US20210158028A1 (en) | Systems and methods for human pose and shape recovery | |
JP4692526B2 (en) | Gaze direction estimation apparatus, gaze direction estimation method, and program for causing computer to execute gaze direction estimation method | |
TWI394093B (en) | An image synthesis method | |
CN108140105A (en) | Head-mounted display with countenance detectability | |
US20050084140A1 (en) | Multi-modal face recognition | |
CN111325846B (en) | Expression base determination method, avatar driving method, device and medium | |
WO2019075666A1 (en) | Image processing method and apparatus, terminal, and storage medium | |
CN103593870A (en) | Picture processing device and method based on human faces | |
CN108388889B (en) | Method and device for analyzing face image | |
CN114219878A (en) | Animation generation method and device for virtual character, storage medium and terminal | |
JP4936491B2 (en) | Gaze direction estimation apparatus, gaze direction estimation method, and program for causing computer to execute gaze direction estimation method | |
CN110796719A (en) | Real-time facial expression reconstruction method | |
CN107749084A (en) | A kind of virtual try-in method and system based on 3-dimensional reconstruction technology | |
WO2021090921A1 (en) | System, program, and method for measuring jaw movement of subject | |
CN112232128B (en) | Eye tracking based method for identifying care needs of old disabled people | |
Lan et al. | The application of 3D morphable model (3DMM) for real-time visualization of acupoints on a smartphone | |
Zhang et al. | 3D statistical head modeling for face/head-related product design: a state-of-the-art review | |
CN108717730B (en) | 3D character reconstruction method and terminal | |
CN114373044A (en) | Method, device, computing equipment and storage medium for generating three-dimensional face model | |
JP4682372B2 (en) | Gaze direction detection device, gaze direction detection method, and program for causing computer to execute gaze direction detection method | |
CN109124765A (en) | Application method of the assisting in diagnosis and treatment system in neurosurgery based on mixed reality |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181228 |