CN110120101A - Cylindrical body augmented reality method, system, device based on 3D vision - Google Patents
Cylindrical body augmented reality method, system, device based on 3D vision Download PDFInfo
- Publication number
- CN110120101A CN110120101A CN201910360629.XA CN201910360629A CN110120101A CN 110120101 A CN110120101 A CN 110120101A CN 201910360629 A CN201910360629 A CN 201910360629A CN 110120101 A CN110120101 A CN 110120101A
- Authority
- CN
- China
- Prior art keywords
- image
- cylindrical body
- augmented reality
- camera posture
- dimensional
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 81
- 230000003190 augmentative effect Effects 0.000 title claims abstract description 79
- 239000012141 concentrate Substances 0.000 claims abstract description 22
- 238000011426 transformation method Methods 0.000 claims abstract description 10
- 239000011159 matrix material Substances 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000013519 translation Methods 0.000 claims description 9
- 239000000284 extract Substances 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 8
- 230000009466 transformation Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 15
- 230000008569 process Effects 0.000 description 9
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 9
- 229910052500 inorganic mineral Inorganic materials 0.000 description 8
- 239000011707 mineral Substances 0.000 description 8
- 239000000571 coke Substances 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000001514 detection method Methods 0.000 description 4
- 239000003550 marker Substances 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/10—Constructive solid geometry [CSG] using solid primitives, e.g. cylinders, cubes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/006—Mixed reality
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- 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
- G06T2207/10016—Video; Image sequence
- G06T2207/10021—Stereoscopic video; Stereoscopic image sequence
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Computer Graphics (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention belongs to computer vision fields, and in particular to a kind of cylindrical body augmented reality method, system, device based on 3D vision, it is intended to solve the problems, such as that the prior art is difficult to carry out augmented reality to cylindrical body.The method of the present invention includes: to concentrate each image to the video image of the cylindrical body multi-angle of view of acquisition, using the contour line of Hough transformation method fitting cylindrical body in the picture, and builds world coordinate system;Based on projective invariance and image-forming principle, the corresponding camera posture of every image is calculated;Reconstruction cylinder threedimensional model;Obtaining three-dimensional model video image based on reconstruction simultaneously carries out the calculating of interframe camera Attitude Tracking, obtains the corresponding camera posture of each frame image;Virtual image is added on cylindrical body video image by the camera posture of acquisition, realizes cylindrical body augmented reality.The method of the present invention is rebuild offline and online tracking accuracy height, speed are fast, and the dummy object being superimposed is stablized, and the purpose of cylindrical body augmented reality is realized.
Description
Technical field
The invention belongs to computer vision fields, and in particular to a kind of cylindrical body augmented reality side based on 3D vision
Method, system, device.
Background technique
In recent years, augmented reality changes the theory of traditional video information interactive application, military in medical treatment, educates, joy
There is huge application prospect in the fields such as happy.Either in academia still in industry, augmented reality is all received widely
Concern.Originally, there are many markers of planar square to be used for augmented reality, such as ARToolKit, ARTag, AprilTag
Deng.ARToolKit be it is earliest most popular, the thought that ARTag and AprilTa are all based on ARToolKit improves
's.Later, the marker of planar rondure gradually came into vogue, such as Mono-spectrum, CCTag, RUNETag etc..No matter
It is the marker of planar square or the marker of planar rondure, is all that dummy object is projected to very according to calculating camera posture
Augmented reality is realized on real object.For these planar tags, their all common disadvantages, it is necessary to it is flat to print these
Then face marker is put the planar tags of printing can realize augmented reality in the scene, and it is not very convenient to operate.Into
One step occurs not needing the augmented reality of label, but is all to be confined in the enhancing to plane, for cylindrical body curved surface
Augmented reality it is difficult very much, the relevant technologies are fresh few at home and abroad.
Summary of the invention
In order to solve the above problem in the prior art, i.e. the prior art is difficult to ask cylindrical body progress augmented reality
Topic, the cylindrical body augmented reality method based on 3D vision that the present invention provides a kind of, comprising:
Step S10 obtains cylindrical body multi-vision angle video image collection, as input picture collection;
Step S20 concentrates each image to the input picture, using Hough transformation method fitting cylindrical body in image
In contour line, and establish the world coordinate system of the Circle in Digital Images cylinder;
Step S30, it is the contour line for the cylindrical body for concentrating each image to be fitted based on the input picture, corresponding
World coordinate system is based on projective invariance and image-forming principle, calculates the corresponding camera posture of every image;
Step S40 extracts the characteristic point that the input picture concentrates each image, and is based on the corresponding phase of described image
Machine posture, the corresponding spatial point of reconstruction features point obtain the cylinder body three-dimensional models of reconstruction;
Step S50, the cylinder body three-dimensional models based on the reconstruction obtain corresponding cylindrical body video image and carry out just
Beginningization obtains initial camera posture and three-dimensional-two-dimentional corresponding relationship;
Step S60 carries out interframe camera Attitude Tracking based on the initial camera posture and three-dimensional-two-dimentional corresponding relationship
It calculates, obtains the corresponding camera posture of each frame image;
The virtual image of input is added to the circle using the corresponding camera posture of each frame image by step S70
On cylinder video image, to realize cylindrical body augmented reality.
In some preferred embodiments, " each image is concentrated to the input picture, using Hough in step S20
Transform method is fitted the contour line of cylindrical body in the picture, and establishes the world coordinate system of the Circle in Digital Images cylinder ", method
Are as follows:
Step S201 concentrates each image to the input picture, using the two of Hough transformation method fitting cylindrical body
Edge line l1And l2, while being fitted two conic section c of cylindrical body1And c2;
Step S202, with the curve c2Central point o2Origin of the corresponding spatial point as world coordinate system, the song
Line c2Central point o2To curve c2On a point X-axis of the straight line as world coordinate system, the curve c2Central point o2
To curve c1Central point o1Z axis of the straight line as world coordinate system, the conic section c2Corresponding space plane is generation
The X-Y plane of boundary's coordinate system completes the foundation of world coordinate system.
In some preferred embodiments, " each image is concentrated to be fitted to obtain based on the input picture in step S30
Cylindrical body contour line, corresponding world coordinate system, be based on projective invariance and image-forming principle, calculate every image it is corresponding
Camera posture ", method are as follows:
Two straight lines and two curves, world coordinate systems based on fitting, calculate separately every image from world coordinate system
Rotational transformation matrix R and translation vector t, the rotational transformation matrix R and translation vector t to camera coordinates system is corresponding for image
Camera posture.
In some preferred embodiments, " feature that the input picture concentrates each image is extracted in step S40
Point, and it is based on the corresponding camera posture of described image, the corresponding spatial point of reconstruction features point " it is additionally provided with spatial point optimization later
The step of, method are as follows:
Step B10 is minimized using re-projection error according to three-dimensional-two-dimensional corresponding relationship and is optimized each frame image
All spatial points observed by posture and image;
Step B20, all spatial points observed by posture and image based on each frame image after the optimization,
Optimize the spatial point of all images and the camera posture of all images using global bundle adjustment.
In some preferred embodiments, " the cylinder body three-dimensional models based on the reconstruction are obtained and are corresponded in step S50
Cylindrical body video image and initialized ", method are as follows:
Step S501, the cylindrical body video image based on acquisition handle image using Linear P3P RANSAC, continuously
Obtain the default corresponding camera posture of frame number image;
Step S502, judge the corresponding camera posture of the default frame number image degree of closeness whether be more than setting threshold
Value, judging result be it is yes, then initialize completion;Judging result be it is no, then follow the steps S501.
In some preferred embodiments, " based on initial camera posture pass corresponding with three-dimensional-two dimension in step S60
System carries out interframe camera Attitude Tracking and calculates, obtains the corresponding camera posture of each frame image ", method are as follows:
Step S601 detects angle point in the area-of-interest of current frame image and extracts two valued description;
Step S602 matches two valued description of current frame image and two valued description of previous frame image, obtains current
Two dimension-two-dimentional relation of frame image and previous frame image;
Step S603, two of three-dimensional-two-dimentional relation and current frame image and previous frame image based on previous frame image
Dimension-two-dimentional relation obtains three-dimensional-two-dimentional relation of current frame image;
Step S604, three-dimensional-two-dimentional relation based on the current frame image calculate current frame image using EPnP method
Corresponding camera posture.
In some preferred embodiments, and in step S70 " the corresponding camera posture of each frame image is used, it will be defeated
The virtual image entered is added on the cylindrical body video image, carries out cylindrical body augmented reality " it is additionally provided with camera appearance before
The step of state unstability is eliminated, method are as follows:
Using extended Kalman filter smooth camera posture, the unstability of camera posture is eliminated.
Another aspect of the present invention proposes a kind of cylindrical body augmented reality system based on 3D vision, including input
At the beginning of module, world coordinate system establish module, camera Attitude Calculation module, cylindrical body three-dimensional reconstruction module, cylindrical body video image
Beginningization module, interframe camera Attitude Tracking module, augmented reality module, output module;
The input module is configured to obtain the video image collection of cylindrical body setting multi-angle of view and input;
The world coordinate system establishes module, is configured to concentrate the video image of the multi-angle of view each image, intends
Cylindrical body contour line is closed, and establishes world coordinate system;
The camera Attitude Calculation module is configured to cylindrical body contour line, the world coordinate system of fitting, utilizes projection
Invariance and image-forming principle calculate the corresponding camera posture of every image;
The cylindrical body three-dimensional reconstruction module, the video image for being configured to extract the multi-angle of view concentrate each image
Characteristic point, and it is based on the corresponding camera posture of described image, the corresponding spatial point of reconstruction features point obtains the cylindrical body three of reconstruction
Dimension module;
The cylindrical body video image initialization module is configured to the cylinder body three-dimensional models of the reconstruction, obtains
Corresponding cylindrical body video image is simultaneously initialized, and initial camera posture and three-dimensional-two-dimentional corresponding relationship are obtained;
The interframe camera Attitude Tracking module is configured to initial camera posture pass corresponding with three-dimensional-two dimension
System carries out interframe camera Attitude Tracking and calculates, obtains the corresponding camera posture of each frame image
The augmented reality module is configured to use the corresponding camera posture of each frame image, by the virtual of input
On image superposition to the cylindrical body video image, cylindrical body augmented reality is carried out;
The output module is configured as output to the later cylindrical body video image of augmented reality.
The third aspect of the present invention proposes a kind of storage device, wherein be stored with a plurality of program, described program be suitable for by
Processor is loaded and is executed to realize the above-mentioned cylindrical body augmented reality method based on 3D vision.
The fourth aspect of the present invention proposes a kind of processing unit, including processor, storage device;The processor is fitted
In each program of execution;The storage device is suitable for storing a plurality of program;Described program be suitable for loaded by processor and executed with
Realize the above-mentioned cylindrical body augmented reality method based on 3D vision.
Beneficial effects of the present invention:
The invention proposes a kind of cylindrical body augmented reality based on 3D vision realizes cylindrical body in off-line phase
Reconstructing three-dimensional model realizes the online three-dimensional tracking of cylindrical body in on-line stage.In actual use, either offline to rebuild also
It is to track online, all obtains very high precision.In addition, online tracking velocity is more than 50FPS, obtained using online tracking
Camera posture is superimposed dummy object, and dummy object is highly stable, realizes the purpose of cylindrical body augmented reality.
Detailed description of the invention
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the application's is other
Feature, objects and advantages will become more apparent upon:
Fig. 1 is the system flow schematic diagram of the cylindrical body augmented reality method the present invention is based on 3D vision;
Fig. 2 is a kind of cylindrical shape of embodiment of the cylindrical body augmented reality method the present invention is based on 3D vision
Object exemplary diagram;
Fig. 3 is that a kind of world of the foundation of embodiment of the cylindrical body augmented reality method the present invention is based on 3D vision is sat
Mark system schematic diagram;
Fig. 4 is a kind of cylinder volume reconstruction mould of embodiment of the cylindrical body augmented reality method the present invention is based on 3D vision
Type exemplary diagram;
Fig. 5 the present invention is based on a kind of camera posture interframe of embodiment of the cylindrical body augmented reality method of 3D vision with
The re-projection error schematic diagram of each frame in track;
The present invention is based on each frames in the tracking of the camera posture interframe of the cylindrical body augmented reality method of 3D vision to disappear by Fig. 6
The time diagram of consumption;
Fig. 7 is the system operation visualization schematic diagram of the cylindrical body augmented reality method the present invention is based on 3D vision;
The present invention is based on a kind of superposition virtual earths of embodiment of the cylindrical body augmented reality method of 3D vision by Fig. 8
Augmented reality result exemplary diagram;
The present invention is based on a kind of replacement cylinder circumferences of embodiment of the cylindrical body augmented reality method of 3D vision by Fig. 9
The augmented reality result exemplary diagram of texture.
Specific embodiment
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is only used for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to just
Part relevant to related invention is illustrated only in description, attached drawing.
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.The application is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
A kind of cylindrical body augmented reality method based on 3D vision of the invention, comprising:
Step S10 obtains cylindrical body multi-vision angle video image collection, as input picture collection;
Step S20 concentrates each image to the input picture, using Hough transformation method fitting cylindrical body in image
In contour line, and establish the world coordinate system of the Circle in Digital Images cylinder;
Step S30, it is the contour line for the cylindrical body for concentrating each image to be fitted based on the input picture, corresponding
World coordinate system is based on projective invariance and image-forming principle, calculates the corresponding camera posture of every image;
Step S40 extracts the characteristic point that the input picture concentrates each image, and is based on the corresponding phase of described image
Machine posture, the corresponding spatial point of reconstruction features point obtain the cylinder body three-dimensional models of reconstruction;
Step S50, the cylinder body three-dimensional models based on the reconstruction obtain corresponding cylindrical body video image and carry out just
Beginningization obtains initial camera posture and three-dimensional-two-dimentional corresponding relationship;
Step S60 carries out interframe camera Attitude Tracking based on the initial camera posture and three-dimensional-two-dimentional corresponding relationship
It calculates, obtains the corresponding camera posture of each frame image;
The virtual image of input is added to the circle using the corresponding camera posture of each frame image by step S70
On cylinder video image, to realize cylindrical body augmented reality.
In order to be more clearly illustrated to the cylindrical body augmented reality method the present invention is based on 3D vision, below with reference to
Fig. 1 is unfolded to be described in detail to step each in embodiment of the present invention method.
The cylindrical body augmented reality method based on 3D vision of an embodiment of the present invention, including step S10- step
S70, each step are described in detail as follows:
Step S10 obtains cylindrical body multi-vision angle video image collection, as input picture collection.
Before the offline three-dimensional reconstruction for carrying out cylindrical body, calibration for cameras Intrinsic Matrix K first utilizes camera intrinsic parameter
Image is normalized in matrix K;Then, some images are shot, comprising cylindrical body and there are multiple visual angles in these images.
As shown in Fig. 2, for the present invention is based on a kind of cylinders of embodiment of the cylindrical body augmented reality method of 3D vision
The object exemplary diagram of shape is from left to right followed successively by can- can, cola- Coke bottle, sprite- Sprite bottle, water-
Mineral water bottle.
Step S20 concentrates each image to the input picture, using Hough transformation method fitting cylindrical body in image
In contour line, and establish the world coordinate system of the Circle in Digital Images cylinder.
The basic principle of Hough transformation method (Hough transform) is the duality using point with line, will be original
The given curve negotiating curve representation form of image space becomes a point of parameter space, thus giving in original image
The test problems for determining curve are converted into the spike problem found in parameter space, namely detection overall permanence is converted into detection office
Portion's characteristic, such as straight line, ellipse, circle, camber line etc..
Step S201 concentrates each image to the input picture, using the two of Hough transformation method fitting cylindrical body
Edge line l1And l2, while being fitted two conic section c of cylindrical body1And c2。
Step S202, with the curve c2Central point o2Origin of the corresponding spatial point as world coordinate system, the song
Line c2Central point o2To curve c2On a point X-axis of the straight line as world coordinate system, the curve c2Central point o2
To curve c1Central point o1Z axis of the straight line as world coordinate system, the conic section c2Corresponding space plane is generation
The X-Y plane of boundary's coordinate system, establishes world coordinate system.
Set the conic section c of fitting1Center be o1, fitting conic section c2Center be o2, choose o2(or o1)
Corresponding spatial point is the origin of world coordinate system, and homogeneous coordinates are (0 00 1)T.In conic section c2One image of upper selection
Point M0, o2To M0Straight line be space coordinates X-axis, then M0Corresponding spatial point homogeneous coordinates are (r 00 1)T, r is cylinder
The radius of body.o2To o1Z axis of the straight line as space coordinates, then o1Corresponding spatial point homogeneous coordinates are (0 0 h 1)T,
H is the height of cylindrical body.Choose conic section c2Corresponding space plane is the X-Y plane of world coordinate system.So far, the world is sat
Mark system is established.
As shown in figure 3, for the present invention is based on a kind of foundation of embodiment of the cylindrical body augmented reality method of 3D vision
World coordinate system schematic diagram, o be world coordinate system origin, o to curve c2On a point M0Straight line be world coordinate system
X-axis, o to curve c1Central point o1Straight line be world coordinate system Z axis, conic section c2Corresponding space plane is
The X-Y plane of world coordinate system.In actual application, world coordinate system is established in this approach, can choose other differences
Origin of the point as world coordinate system, also can choose the X, Y, Z axis of straight line that other pass through origin as world coordinate system,
This is no longer going to repeat them.
Step S30, it is the contour line for the cylindrical body for concentrating each image to be fitted based on the input picture, corresponding
World coordinate system is based on projective invariance and image-forming principle, calculates the corresponding camera posture of every image.
Common projective invariance includes that the projection of collinear points is still conllinear, and the projection of parallel lines intersects at a point, directly
The cross ratio invariability etc. of line segment projection.
Cylindrical body contour line (including two straight lines and two curves), world coordinate system based on fitting calculate separately every
Open spin matrix R and translation vector t, the rotational transformation matrix R and translation of the image from world coordinate system to camera coordinates system
Vector t is the corresponding camera posture of image.
Rotational transformation matrix R and translation vector t, as shown in formula (1) and formula (2):
R=(r1 r2 r3) formula (1)
T=(t1 t2 t3) formula (2)
Wherein, r1=(r11 r21 r31)T, r2=(r12 r22 r32)T, r3=(r13 r23 r33)TIt is the three of spin matrix R
Column.
Infinite point V in world coordinate system, on Z axisz=(0 01 0)T, define fitting a straight line l1And l2Intersection point
For VzIn the subpoint v of two dimensional imagez.By the image-forming principle of image, v is calculatedz, as shown in formula (3):
vz=l1×l2≈(r1 r2 r3 t)(0 0 1 0)TFormula (3)
So as to calculate r3, as shown in formula (4):
According to projective invariance, digital simulation curve c1Central point o1, matched curve c2Central point o2, such as formula (5) and
Shown in formula (6):
Wherein, u1、v1Respectively represent o1Abscissa, ordinate in two dimensional image, u2、v2Respectively represent o2In two dimensional image
Abscissa, ordinate.
According to o1The homogeneous coordinates (0 0 h 1) of corresponding spatial pointTAnd o2The homogeneous coordinates (0 00 1) of corresponding spatial pointT,
Based on image image-forming principle, scale factor s1、s2Calculation method such as formula (7) and formula (8) shown in:
So as to calculate t, as shown in formula (9):
T=s1*o1-h*r3Formula (9)
Under world coordinate system, the infinite point homogeneous coordinates of X-axis are (1 00 1)T, according to image image-forming principle, meter
Calculate vx, as shown in formula (10):
vx=(o2×m0)×Vz≈(r1 r2 r3 t)(1 0 0 0)TFormula (10)
So as to calculate r1, as shown in formula (11):
Calculate r2, as shown in formula (12):
r2=r3×r1Formula (12)
In conclusion obtaining rotational transformation matrix R and translation vector t is the corresponding camera posture of image.
Step S40 extracts the characteristic point that the input picture concentrates each frame image, and is based on the corresponding phase of described image
Machine posture, the corresponding spatial point of reconstruction features point obtain the cylinder body three-dimensional models of reconstruction.
On the first frame image of the input picture collection, a characteristic point m is choseni=(ui vi 1)T, corresponding sky
Between point homogeneous coordinates be Mi=(Xi Yi Zi 1)T, then according to imaging process, MiShown in the calculation method of reconstruction such as formula (13):
(Xi Yi Zi)T=s*RT*mi-RT* t formula (13)
Wherein, s is scale factor;Spatial point MiOn cylindrical body, meetR is cylinder radius.
" characteristic point that the input picture concentrates each image is extracted, and corresponding based on described image in step S40
The step of spatial point optimization is additionally provided with after camera posture, the corresponding spatial point of reconstruction features point ", method are as follows:
Step B10 is minimized using re-projection error according to three-dimensional-two-dimensional corresponding relationship and is optimized each frame image
All spatial points observed by posture and image, as shown in formula (14):
Step B20, all spatial points observed by posture and image based on each frame image after the optimization,
Optimize the spatial point of all images and the camera posture of all images using global bundle adjustment, as shown in formula (15):
Wherein, KlRepresent all images, PlAll spatial points are represented, E (i, j) is represented, calculation method such as formula (16) institute
Show:
E (i, j)=| | mi-[Rj,tj]Mi||2Formula (16)
As shown in figure 4, for the present invention is based on a kind of cylinders of embodiment of the cylindrical body augmented reality method of 3D vision
Volume reconstruction model example figure, upper left are can- can, and upper right is cola- Coke bottle, and lower-left is sprite- Sprite bottle, bottom right
For water- mineral water bottle.
Cylindrical body three-dimensional reconstruction is carried out using the method for the present invention, precision is high, error is low, and be averaged reconstruction error such as 1 institute of table
Show:
Table 1
Cylindrical body | Can | Coke bottle | Sprite bottle | Mineral water bottle |
Average reconstruction error (pixel) | 1.92×10-5 | 1.66×10-5 | 1.60×10-5 | 1.66×10-5 |
Step S50, the cylindrical body 3-D image based on the reconstruction obtain corresponding cylindrical body video image and carry out just
Beginningization obtains initial camera posture and three-dimensional-two-dimentional corresponding relationship.
Pose solution is frequently encountered in computer vision, and P3P (P3P, Perspective-3-Points) is provided
A solution, it is a kind of pose solution mode by 3D-2D, needs known matched 3D point and image 2D point.
Step S501, the cylindrical body video image based on acquisition handle image using Linear P3P RANSAC, continuously
Obtain the default corresponding camera posture of frame number image.
Step S502, judge the corresponding camera posture of the default frame number image degree of closeness whether be more than setting threshold
Value, judging result be it is yes, then initialize completion;Judging result be it is no, then follow the steps S501.
In the initialization procedure of cylindrical body, Linear P3P RANSAC is used:
Firstly, rejecting the straight line and short straight line close to image border using Hough transformation detection and fitting a straight line, merge
Identical straight line, with set Si={ li, i=1,2 ... N } and indicate remaining straight line.In addition, by calculating between two straight lines
Angle, finds many parallel lines pair, and note is parallel to straight line liStraight line collection be combined into STi.In set STiIn, selection distance is farthest
Two straight lines, reject other straight lines, remember that the intersection point of two farthest straight lines is v, then the third column of spin matrix R can be by
It calculates, as shown in formula (17):
On the image extracted region angle point folded by two farthest straight lines and with two valued description describe these extraction
Angle point.
Then, present frame is found by description corresponding to description of matching present frame and the spatial point rebuild
3D-2D corresponding points remember that the collection of corresponding points is combined into Si={ (mj,Mj), j=1,2 ... n }.From set SiIn arbitrarily select three couples of 3D-
2D corresponding points, are denoted as (mi,Mi), i=1,2,3, according to imaging process, our available formulas (18):
si*mi=RMi+ t formula (18)
Wherein, siIt is scale factor, Mi=(Xi,Yi,Zi)T, i=1,2,3.
Scale factor s can be used1Indicate scale factor s2、s3, as shown in formula (19) and formula (20):
Equation group further can be obtained, as shown in formula (21):
Formula (19) and formula (20) are substituted into equation group (21), obtain one about s1Equation group, use SVD point
Solution solves equation group, the s that then will be solved1Formula (19) and formula (20) are substituted into, s can be calculated2And s3.It will calculate
Obtained s1、s2And s3It substitutes into formula (18), linear solution can obtain r1、r2, as shown in formula (22) and formula (23):
Wherein, A1=(s2m2-s1m1-(Z2-Z1)r3)(Y3-Y1), A2=(s3m3-s1m1-(Z3-Z1)r3)(Y2-Y1), A3=
(s2m2-s1m1-(Z2-Z1)r3)(X3-X1), A4=(s3m3-s1m1-(Z3-Z1)r3)(X2-X1), B==(X2-X1)(Y3-Y1)-
(X3-X1)(Y2-Y1)。
Thus the spin matrix R=(r obtained1 r2 r3), translation vector t is calculated, as shown in formula (24):
T=simi-RMi, i=1,2,3 formula (24)
By camera the posture R and t acquired above, we can calculate interior point number and put number in saving.Again
Three pairs of different 3D-2D corresponding points are selected, process above is repeated and finds interior number and save interior point number.Institute is handled
After the different three couple 3D-2D corresponding points having, camera posture corresponding to interior the largest number of combinations of point is selected.In order to further
Accurate camera posture is obtained, three pairs of 3D-2D corresponding points and Qi Nei point are put together, recalculate camera appearance using EPnP
State.
Finally, in each set STiIn, according to above method, calculate camera posture and it is corresponding in point, in selection
Camera posture corresponding to the largest number of set of point.Then, we advanced optimize camera posture according to corresponding interior point, such as
Shown in formula (25):
Above method, we term it " Linear P3P RANSAC "." Linear P3P RANSAC " is used to handle
Continuous three frames image, if their camera posture is relatively, then it is assumed that initialize successfully, otherwise initialization failure, weight
Continuous three frames image is newly selected to be initialized until initializing successfully.
Step S60 carries out interframe camera Attitude Tracking based on the initial camera posture and three-dimensional-two-dimentional corresponding relationship
It calculates, obtains the corresponding camera posture of each frame image.
Step S601 detects angle point in the area-of-interest of current frame image and extracts two valued description.
Step S602, matches the characteristic point of current frame image and the characteristic point of previous frame image, obtain current frame image with
Two dimension-two-dimentional relation of previous frame image.
Step S603, two of three-dimensional-two-dimentional relation and current frame image and previous frame image based on previous frame image
Dimension-two-dimentional relation obtains three-dimensional-two-dimentional relation of current frame image.
Step S604, three-dimensional-two-dimentional relation based on the current frame image calculate current frame image using EPnP method
Corresponding camera posture.
After initializing successfully, interframe tracking is carried out using the information of previous frame.Interframe tracking include tracking previous frame and with
Track model.During tracking previous frame, firstly, present frame area-of-interest (region where cylindrical body) detection angle point and
Extract description;Then, the sub 2D-2D to find present frame and previous frame of description of description and previous frame of present frame is matched
Corresponding relationship, to find the 3D-2D corresponding relationship of present frame;Finally, being used according to the 3D-2D corresponding relationship of present frame
The camera posture of EPnP calculating present frame.If tracking previous frame failure, the camera of present frame is predicted using motion model
Posture.During trace model, pass through formula s*mi=R (Xi Yi Zi)T+ t and formulaIt rebuilds in present frame
Not with the matched characteristic point of previous frame, the spatial point newly rebuild is matched with model further, obtains more 3D-2D pairs
Ying Dian, using the camera posture of all 3D-2D corresponding points optimization present frames of present frame, as shown in formula (26):
Finally according to the camera posture after optimization, the area-of-interest of next frame is predicted.
As shown in figure 5, for the present invention is based on a kind of cameras of embodiment of the cylindrical body augmented reality method of 3D vision
The re-projection error schematic diagram of each frame, it is laughable from left to right to respectively represent can- can, cola- in the tracking of posture interframe
Bottle, sprite- Sprite bottle, water- mineral water bottle, Frames represent all picture frames in a video, reprojection
Error (pixel) represents re-projection error, and the re-projection error that straight line represents most of picture frames in video is less than a certain setting
Threshold value.
As shown in fig. 6, for the present invention is based on the tracking of the camera posture interframe of the cylindrical body augmented reality method of 3D vision
In the consumption of each frame time diagram, from left to right respectively represent can- can, cola- Coke bottle, sprite- Sprite
Bottle, water- mineral water bottle, Frames represent all picture frames in a video, and times (ms) represents online tracking process
In the consumption of each frame time, straight line, which represents most of picture frames and tracks time of consumption online, is less than a certain given threshold.
Augmented reality is carried out using the method for the present invention, frame loss rate is low during line camera Attitude Tracking, re-projection error
The number of image frames FPS high that small, each second can track online, as shown in table 2:
Video | Frame number | Frame loss rate | Re-projection error (pixel) | FPS(Hz) |
Can | 3356 | 0.70% | 0.90 | 63 |
Cola | 3507 | 1.64% | 1.26 | 59 |
Sprite | 3386 | 1.27% | 0.93 | 63 |
Water | 3449 | 0.19% | 0.95 | 56 |
The virtual image of input is added to the circle using the corresponding camera posture of each frame image by step S70
On cylinder video image, to realize cylindrical body augmented reality.
In step S70 " the corresponding camera posture of each frame image is used, the virtual image of input is added to described
On cylindrical body video image, to realize cylindrical body augmented reality " it is additionally provided with the step of camera unsteady attitude is eliminated before,
Its method are as follows:
Using extended Kalman filter smooth camera posture, the unstability of camera posture is eliminated.
When due to shooting video, the shake that equipment has in a way causes the dummy object of superposition to there may be certain
It is unstable in kind degree, use extended Kalman filter smooth camera posture.It then, will using smoothed out camera posture
Dummy object projects on true cylindrical body, it is ensured that dummy object is stable to be added on cylindrical body, to reach increasing
The purpose of strong reality.
As shown in fig. 7, for the present invention is based on the operation visualizations of the system of the cylindrical body augmented reality method of 3D vision to show
It is intended to, Fig. 7 left figure is that the virtual earth is added on the image containing mineral water bottle, and Fig. 7 right figure is the offline Three-dimensional Gravity of cylindrical body
The pyramid of the model built, the right figure upper left corner represents the spatial position of camera.
As shown in figure 8, for the present invention is based on a kind of superpositions of embodiment of the cylindrical body augmented reality method of 3D vision
The augmented reality result exemplary diagram of virtual earth, from left to right respectively represents virtual earth being superimposed upon can- can, cola-
The effect of Coke bottle, sprite- Sprite bottle, water- mineral water bottle.
As shown in figure 9, for the present invention is based on a kind of replacements of embodiment of the cylindrical body augmented reality method of 3D vision
The augmented reality result exemplary diagram of cylinder circumference texture from left to right respectively represents replacement can- can, cola- is laughable
Bottle, sprite- Sprite bottle, water- mineral water bottle surrounding's texture effect.
The cylindrical body augmented reality system based on 3D vision of second embodiment of the invention, including input module, the world
Establishment of coordinate system module, camera Attitude Calculation module, cylindrical body three-dimensional reconstruction module, cylindrical body video image initialization module,
Interframe camera Attitude Tracking module, augmented reality module, output module;
The input module is configured to obtain the video image collection of cylindrical body setting multi-angle of view and input;
The world coordinate system establishes module, is configured to concentrate the video image of the multi-angle of view each image, intends
Cylindrical body contour line is closed, and establishes world coordinate system;
The camera Attitude Calculation module is configured to cylindrical body contour line, the world coordinate system of fitting, utilizes projection
Invariance and image-forming principle calculate the corresponding camera posture of every image;
The cylindrical body three-dimensional reconstruction module, the video image for being configured to extract the multi-angle of view concentrate each image
Characteristic point, and it is based on the corresponding camera posture of described image, the corresponding spatial point of reconstruction features point obtains the cylindrical body three of reconstruction
Dimension module;
The cylindrical body video image initialization module is configured to the cylinder body three-dimensional models of the reconstruction, obtains
Corresponding cylindrical body video image is simultaneously initialized, and initial camera posture and three-dimensional-two-dimentional corresponding relationship are obtained;
The interframe camera Attitude Tracking module is configured to initial camera posture pass corresponding with three-dimensional-two dimension
System carries out interframe camera Attitude Tracking and calculates, obtains the corresponding camera posture of each frame image
The augmented reality module is configured to use the corresponding camera posture of each frame image, by the virtual of input
On image superposition to the cylindrical body video image, cylindrical body augmented reality is carried out;
The output module is configured as output to the later cylindrical body video image of augmented reality.
Person of ordinary skill in the field can be understood that, for convenience and simplicity of description, foregoing description
The specific work process of system and related explanation, can refer to corresponding processes in the foregoing method embodiment, details are not described herein.
It should be noted that the cylindrical body augmented reality system provided by the above embodiment based on 3D vision, only more than
The division for stating each functional module carries out for example, in practical applications, can according to need and by above-mentioned function distribution by not
With functional module complete, i.e., by the embodiment of the present invention module or step decompose or combine again, for example, above-mentioned reality
The module for applying example can be merged into a module, can also be further split into multiple submodule, described above complete to complete
Portion or partial function.For module involved in the embodiment of the present invention, the title of step, it is only for distinguish modules
Or step, it is not intended as inappropriate limitation of the present invention.
A kind of storage device of third embodiment of the invention, wherein being stored with a plurality of program, described program is suitable for by handling
Device is loaded and is executed to realize the above-mentioned cylindrical body augmented reality method based on 3D vision.
A kind of processing unit of fourth embodiment of the invention, including processor, storage device;Processor is adapted for carrying out each
Program;Storage device is suitable for storing a plurality of program;Described program is suitable for being loaded by processor and being executed to realize above-mentioned base
In the cylindrical body augmented reality method of 3D vision.
Person of ordinary skill in the field can be understood that, for convenience and simplicity of description, foregoing description
The specific work process and related explanation of storage device, processing unit, can refer to corresponding processes in the foregoing method embodiment,
Details are not described herein.
Those skilled in the art should be able to recognize that, mould described in conjunction with the examples disclosed in the embodiments of the present disclosure
Block, method and step, can be realized with electronic hardware, computer software, or a combination of the two, software module, method and step pair
The program answered can be placed in random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electric erasable and can compile
Any other form of storage well known in journey ROM, register, hard disk, moveable magnetic disc, CD-ROM or technical field is situated between
In matter.In order to clearly demonstrate the interchangeability of electronic hardware and software, in the above description according to function generally
Describe each exemplary composition and step.These functions are executed actually with electronic hardware or software mode, depend on technology
The specific application and design constraint of scheme.Those skilled in the art can carry out using distinct methods each specific application
Realize described function, but such implementation should not be considered as beyond the scope of the present invention.
Term " includes " or any other like term are intended to cover non-exclusive inclusion, so that including a system
Process, method, article or equipment/device of column element not only includes those elements, but also including being not explicitly listed
Other elements, or further include the intrinsic element of these process, method, article or equipment/devices.
So far, it has been combined preferred embodiment shown in the drawings and describes technical solution of the present invention, still, this field
Technical staff is it is easily understood that protection scope of the present invention is expressly not limited to these specific embodiments.Without departing from this
Under the premise of the principle of invention, those skilled in the art can make equivalent change or replacement to the relevant technologies feature, these
Technical solution after change or replacement will fall within the scope of protection of the present invention.
Claims (10)
1. a kind of cylindrical body augmented reality method based on 3D vision characterized by comprising
Step S10 obtains cylindrical body multi-vision angle video image collection, as input picture collection;
Step S20 concentrates each image to the input picture, in the picture using Hough transformation method fitting cylindrical body
Contour line, and establish the world coordinate system of the Circle in Digital Images cylinder;
Step S30, the contour line for the cylindrical body for concentrating each image to be fitted based on the input picture, the corresponding world
Coordinate system is based on projective invariance and image-forming principle, calculates the corresponding camera posture of every image;
Step S40 extracts the characteristic point that the input picture concentrates each image, and is based on the corresponding camera appearance of described image
State, the corresponding spatial point of reconstruction features point obtain the cylinder body three-dimensional models of reconstruction;
Step S50, the cylinder body three-dimensional models based on the reconstruction obtain corresponding cylindrical body video image and carry out initial
Change, obtains initial camera posture and three-dimensional-two-dimentional corresponding relationship;
Step S60 is carried out interframe camera Attitude Tracking and is calculated based on the initial camera posture and three-dimensional-two-dimentional corresponding relationship,
Obtain the corresponding camera posture of each frame image;
The virtual image of input is added to the cylindrical body using the corresponding camera posture of each frame image by step S70
On video image, to realize cylindrical body augmented reality.
2. the cylindrical body augmented reality method according to claim 1 based on 3D vision, which is characterized in that step S20
In " each image is concentrated to the input picture, using the contour line of Hough transformation method fitting cylindrical body in the picture, and
Establish the world coordinate system of the Circle in Digital Images cylinder ", method are as follows:
Step S201 concentrates each image to the input picture, using two sides of Hough transformation method fitting cylindrical body
Edge straight line l1And l2, while being fitted two conic section c of cylindrical body1And c2;
Step S202, with the curve c2Central point o2Origin of the corresponding spatial point as world coordinate system, the curve c2
Central point o2To curve c2On a point X-axis of the straight line as world coordinate system, the curve c2Central point o2To song
Line c1Central point o1Z axis of the straight line as world coordinate system, the conic section c2Corresponding space plane is world's seat
The X-Y plane for marking system, completes the foundation of world coordinate system.
3. the cylindrical body augmented reality method according to claim 1 based on 3D vision, which is characterized in that step S30
In " contour line, corresponding world coordinate system based on the cylindrical body that the input picture concentrates each image to be fitted, base
In projective invariance and image-forming principle, the corresponding camera posture of every image is calculated ", method are as follows:
Two straight lines and two curves, world coordinate systems based on fitting, calculate separately every image from world coordinate system to phase
The rotational transformation matrix R and translation vector t, the rotational transformation matrix R and translation vector t of machine coordinate system are the corresponding phase of image
Machine posture.
4. the cylindrical body augmented reality method according to claim 1 based on 3D vision, which is characterized in that step S40
In " characteristic point that the input picture concentrates each image is extracted, and is based on the corresponding camera posture of described image, reconstruction spy
The step of being additionally provided with spatial point optimization after the corresponding spatial point of sign point ", method are as follows:
Step B10 minimizes the posture for optimizing each frame image using re-projection error according to three-dimensional-two-dimensional corresponding relationship
With spatial point all observed by image;
Step B20, all spatial points observed by posture and image based on each frame image after the optimization, uses
Global bundle adjustment optimizes the spatial point of all images and the camera posture of all images.
5. the cylindrical body augmented reality method according to claim 1 based on 3D vision, which is characterized in that step S50
In " the cylinder body three-dimensional models based on the reconstruction obtain corresponding cylindrical body video image and are initialized ", method
Are as follows:
Step S501, the cylindrical body video image based on acquisition handle image using Linear P3P RANSAC, continuous to obtain
The default corresponding camera posture of frame number image;
Step S502, judge the corresponding camera posture of the default frame number image degree of closeness whether be more than setting threshold value,
Judging result be it is yes, then initialize completion;Judging result be it is no, then follow the steps S501.
6. the cylindrical body augmented reality method according to claim 1 based on 3D vision, which is characterized in that step S60
In " based on the initial camera posture and three-dimensional-two-dimentional corresponding relationship, carry out interframe camera Attitude Tracking and calculate, obtain each
The corresponding camera posture of frame image ", method are as follows:
Step S601 detects angle point in the area-of-interest of current frame image and extracts two valued description;
Step S602 matches two valued description of current frame image and two valued description of previous frame image, obtains present frame figure
As two dimension-two-dimentional relation with previous frame image;
Step S603, the two dimension-two of three-dimensional-two-dimentional relation and current frame image and previous frame image based on previous frame image
Dimension relationship obtains three-dimensional-two-dimentional relation of current frame image;
It is corresponding to calculate current frame image using EPnP method for step S604, three-dimensional-two-dimentional relation based on the current frame image
Camera posture.
7. the cylindrical body augmented reality method according to claim 1 based on 3D vision, which is characterized in that step S70
In " using the corresponding camera posture of each frame image, the virtual image of input is added to the cylindrical body video image
On, carry out cylindrical body augmented reality " it is additionally provided with the step of camera unsteady attitude is eliminated, method before are as follows:
Using extended Kalman filter smooth camera posture, the unstability of camera posture is eliminated.
8. a kind of cylindrical body augmented reality system based on 3D vision, which is characterized in that including input module, world coordinate system
Establish module, camera Attitude Calculation module, cylindrical body three-dimensional reconstruction module, cylindrical body video image initialization module, interframe phase
Machine Attitude Tracking module, augmented reality module, output module;
The input module is configured to obtain the video image collection of cylindrical body setting multi-angle of view and input;
The world coordinate system establishes module, is configured to concentrate the video image of the multi-angle of view each image, fitting circle
Cylinder contour line, and establish world coordinate system;
The camera Attitude Calculation module is configured to cylindrical body contour line, the world coordinate system of fitting, constant using projection
Property and image-forming principle, calculate the corresponding camera posture of every image;
The cylindrical body three-dimensional reconstruction module, the video image for being configured to extract the multi-angle of view concentrate the feature of each image
Point, and it is based on the corresponding camera posture of described image, the corresponding spatial point of reconstruction features point obtains the cylindrical body three-dimensional mould of reconstruction
Type;
The cylindrical body video image initialization module is configured to the cylinder body three-dimensional models of the reconstruction, obtains and corresponds to
Cylindrical body video image and initialized, obtain initial camera posture and three-dimensional-two-dimentional corresponding relationship;
The interframe camera Attitude Tracking module is configured to the initial camera posture and three-dimensional-two-dimentional corresponding relationship, into
Row interframe camera Attitude Tracking calculates, and obtains the corresponding camera posture of each frame image
The augmented reality module is configured to use the corresponding camera posture of each frame image, by the virtual image of input
It is added on the cylindrical body video image, carries out cylindrical body augmented reality;
The output module is configured as output to the later cylindrical body video image of augmented reality.
9. a kind of storage device, wherein being stored with a plurality of program, which is characterized in that described program is suitable for being loaded and being held by processor
Row is to realize the described in any item cylindrical body augmented reality methods based on 3D vision of claim 1-8.
10. a kind of processing unit, including
Processor is adapted for carrying out each program;And
Storage device is suitable for storing a plurality of program;
It is characterized in that, described program is suitable for being loaded by processor and being executed to realize:
The described in any item cylindrical body augmented reality methods based on 3D vision of claim 1-8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910360629.XA CN110120101B (en) | 2019-04-30 | 2019-04-30 | Cylinder augmented reality method, system and device based on three-dimensional vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910360629.XA CN110120101B (en) | 2019-04-30 | 2019-04-30 | Cylinder augmented reality method, system and device based on three-dimensional vision |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110120101A true CN110120101A (en) | 2019-08-13 |
CN110120101B CN110120101B (en) | 2021-04-02 |
Family
ID=67520319
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910360629.XA Active CN110120101B (en) | 2019-04-30 | 2019-04-30 | Cylinder augmented reality method, system and device based on three-dimensional vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110120101B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112288823A (en) * | 2020-10-15 | 2021-01-29 | 武汉工程大学 | Calibration method of standard cylinder curved surface point measuring equipment |
CN112734914A (en) * | 2021-01-14 | 2021-04-30 | 温州大学 | Image stereo reconstruction method and device for augmented reality vision |
CN113673283A (en) * | 2020-05-14 | 2021-11-19 | 惟亚(上海)数字科技有限公司 | Smooth tracking method based on augmented reality |
CN114549660A (en) * | 2022-02-23 | 2022-05-27 | 北京大学 | Multi-camera calibration method, device and equipment based on cylindrical self-identification marker |
CN115115708A (en) * | 2022-08-22 | 2022-09-27 | 荣耀终端有限公司 | Image pose calculation method and system |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101101505A (en) * | 2006-07-07 | 2008-01-09 | 华为技术有限公司 | Method and system for implementing three-dimensional enhanced reality |
KR20170090165A (en) * | 2016-01-28 | 2017-08-07 | 허상훈 | Apparatus for realizing augmented reality using multiple projector and method thereof |
CN108898630A (en) * | 2018-06-27 | 2018-11-27 | 清华-伯克利深圳学院筹备办公室 | A kind of three-dimensional rebuilding method, device, equipment and storage medium |
US20190026948A1 (en) * | 2017-07-24 | 2019-01-24 | Visom Technology, Inc. | Markerless augmented reality (ar) system |
US20190051054A1 (en) * | 2017-08-08 | 2019-02-14 | Smart Picture Technologies, Inc. | Method for measuring and modeling spaces using markerless augmented reality |
CN109389634A (en) * | 2017-08-02 | 2019-02-26 | 蒲勇飞 | Virtual shopping system based on three-dimensional reconstruction and augmented reality |
CN109472873A (en) * | 2018-11-02 | 2019-03-15 | 北京微播视界科技有限公司 | Generation method, device, the hardware device of threedimensional model |
CN109685913A (en) * | 2018-12-21 | 2019-04-26 | 西安电子科技大学 | Augmented reality implementation method based on computer vision positioning |
-
2019
- 2019-04-30 CN CN201910360629.XA patent/CN110120101B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101101505A (en) * | 2006-07-07 | 2008-01-09 | 华为技术有限公司 | Method and system for implementing three-dimensional enhanced reality |
KR20170090165A (en) * | 2016-01-28 | 2017-08-07 | 허상훈 | Apparatus for realizing augmented reality using multiple projector and method thereof |
US20190026948A1 (en) * | 2017-07-24 | 2019-01-24 | Visom Technology, Inc. | Markerless augmented reality (ar) system |
CN109389634A (en) * | 2017-08-02 | 2019-02-26 | 蒲勇飞 | Virtual shopping system based on three-dimensional reconstruction and augmented reality |
US20190051054A1 (en) * | 2017-08-08 | 2019-02-14 | Smart Picture Technologies, Inc. | Method for measuring and modeling spaces using markerless augmented reality |
CN108898630A (en) * | 2018-06-27 | 2018-11-27 | 清华-伯克利深圳学院筹备办公室 | A kind of three-dimensional rebuilding method, device, equipment and storage medium |
CN109472873A (en) * | 2018-11-02 | 2019-03-15 | 北京微播视界科技有限公司 | Generation method, device, the hardware device of threedimensional model |
CN109685913A (en) * | 2018-12-21 | 2019-04-26 | 西安电子科技大学 | Augmented reality implementation method based on computer vision positioning |
Non-Patent Citations (1)
Title |
---|
ASAHI SUZUKI 等: "Design of an AR Marker for Cylindrical Surface", 《2013 IEEE INTERNATIONAL SYMPOSIUM ON MIXED AND AUGMENTED REALITY ISMAR》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113673283A (en) * | 2020-05-14 | 2021-11-19 | 惟亚(上海)数字科技有限公司 | Smooth tracking method based on augmented reality |
CN112288823A (en) * | 2020-10-15 | 2021-01-29 | 武汉工程大学 | Calibration method of standard cylinder curved surface point measuring equipment |
CN112734914A (en) * | 2021-01-14 | 2021-04-30 | 温州大学 | Image stereo reconstruction method and device for augmented reality vision |
CN114549660A (en) * | 2022-02-23 | 2022-05-27 | 北京大学 | Multi-camera calibration method, device and equipment based on cylindrical self-identification marker |
CN114549660B (en) * | 2022-02-23 | 2022-10-21 | 北京大学 | Multi-camera calibration method, device and equipment based on cylindrical self-identification marker |
CN115115708A (en) * | 2022-08-22 | 2022-09-27 | 荣耀终端有限公司 | Image pose calculation method and system |
Also Published As
Publication number | Publication date |
---|---|
CN110120101B (en) | 2021-04-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110120101A (en) | Cylindrical body augmented reality method, system, device based on 3D vision | |
CN109214282B (en) | A kind of three-dimension gesture critical point detection method and system neural network based | |
CN105913489B (en) | A kind of indoor three-dimensional scenic reconstructing method using plane characteristic | |
CN104835144B (en) | The method for solving camera intrinsic parameter using the picture and orthogonality of the centre of sphere of a ball | |
CN104748746B (en) | Intelligent machine attitude determination and virtual reality loaming method | |
Tang et al. | 3D mapping and 6D pose computation for real time augmented reality on cylindrical objects | |
CN102663820B (en) | Three-dimensional head model reconstruction method | |
CN103729885B (en) | Various visual angles projection registers united Freehandhand-drawing scene three-dimensional modeling method with three-dimensional | |
CN109272537A (en) | A kind of panorama point cloud registration method based on structure light | |
CN106503671A (en) | The method and apparatus for determining human face posture | |
CN109035327B (en) | Panoramic camera attitude estimation method based on deep learning | |
CN102750704B (en) | Step-by-step video camera self-calibration method | |
CN109523589A (en) | A kind of design method of more robust visual odometry | |
CN107677274A (en) | Unmanned plane independent landing navigation information real-time resolving method based on binocular vision | |
CN103077546B (en) | The three-dimensional perspective transform method of X-Y scheme | |
CN107886546A (en) | Utilize the method for ball picture and public self-polar triangle demarcation parabolic catadioptric video camera | |
CN107145224B (en) | Human eye sight tracking and device based on three-dimensional sphere Taylor expansion | |
CN106485207A (en) | A kind of Fingertip Detection based on binocular vision image and system | |
Petit et al. | A robust model-based tracker combining geometrical and color edge information | |
CN108492017A (en) | A kind of product quality information transmission method based on augmented reality | |
CN104537705A (en) | Augmented reality based mobile platform three-dimensional biomolecule display system and method | |
CN107194984A (en) | Mobile terminal real-time high-precision three-dimensional modeling method | |
CN110349225A (en) | A kind of BIM model exterior contour rapid extracting method | |
CN101196988B (en) | Palm locating and center area extraction method of three-dimensional palm print identity identification system | |
CN115830135A (en) | Image processing method and device and electronic equipment |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |