CN1567385A - Panoramic reconstruction method of three dimensional image from two dimensional image - Google Patents
Panoramic reconstruction method of three dimensional image from two dimensional image Download PDFInfo
- Publication number
- CN1567385A CN1567385A CN 03137660 CN03137660A CN1567385A CN 1567385 A CN1567385 A CN 1567385A CN 03137660 CN03137660 CN 03137660 CN 03137660 A CN03137660 A CN 03137660A CN 1567385 A CN1567385 A CN 1567385A
- Authority
- CN
- China
- Prior art keywords
- image
- depth
- anaglyph
- picture
- pixel
- 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
Landscapes
- Processing Or Creating Images (AREA)
Abstract
This invention discloses a kind of method to reconstruct the panoramic stereo picture from flat picture. The course comprises: selecting flat picture; building the depth list of every pixel aiming at the space distribution of flat picture; executing the parallactic displacement process to every pixel in depth list; reconstructing all the parallax sequence image, then stereo synthesizing the parallax sequence image. The panoramic reconstruction method can convert the flat picture to vivid stereo picture. The stereo depth of field can be controlled at will. It expands the applied field of stereo picture.
Description
The invention belongs to optics display technique field, relate to a kind of method of reconstruction image plane stereo-picture.
Along with reaching its maturity of grating stereo imaging technique, the 3 D stereo picture watches freedom, characteristics clear, that stereoscopic sensation is strong to begin to be popularized with it.The graphical design of 3 D stereo picture is called for short three-dimensional design, and up to the present, main stereoscopic photograph, the three-dimensional modeling of adopting of three-dimensional design played up technology or figure layer translation technology from various visual angles.
Stereoscopic photograph need be taken the model at a plurality of parallax points, particularly must adopt stereoscopic camera to non-still life model such as personage, animal etc., because professional stereoscopic camera volume is bigger, poor operation convenience, be unfavorable for the candid photograph of splendid moment, shooting effect is all the time not as the photography of professional plane, and therefore many photographers are reluctant initiatively to use the stereoscopic shooting means, have hindered stereoscopic photograph in the coml widespread use.
Three-dimensional modeling is played up Technology Need from various visual angles and is set up stereoscopic model, pinup picture is also arranged light, camera is set to be played up model at a plurality of parallax points, obtain several parallax sequence images, can obtain extraordinary stereoeffect, but be difficult for making real stereo scene in the life, there is certain limitation in its utilization, and this method in fact also should belong to stereographic category.
Figure layer translation technology is most widely used in current solid design, its essence is a sheet of planar image is reached the purpose that the space stretches through the layering skew, because it is simple, nearly all designer all with it as main design means, shortcoming is, its stereoscopic sensation is a planar layer, lacks space continuity.Although can increase each layer edges seamlessly transitting on the degree of depth, eliminate the sheet sense of main body, prospect, background, perhaps by cut sth. askew, horizontal compression will scheme layer and stretch at any direction in space, further improve the designing quality of figure layer translation technology, but carry out depth localization because its design concept is based on to figure layer " face " rather than to space " point ", can't design the three dimensions sense as stereoscopic photograph.
Plane picture panorama disclosed by the invention is rebuild the method for stereo-picture, combines dimensional Modeling Technology, image geometry converter technique and image parallactic converter technique, and purpose provides a kind of method that the plane picture panorama is reconstructed into the high precision stereo-picture.
The principle that panorama is rebuild is, at the space distribution of plane picture, set up the degree of depth tabulation of each pixel of image, according to degree of depth tabulation each pixel of plane picture is carried out parallaxial displacement and handle, obtain several parallax sequence images, then that the parallax sequence image is three-dimensional synthetic.Although the degree of depth tabulation of being set up by plane picture can not reflect the three-dimensional depth of image reality fully truly, consider that there is strong degree of depth psychological hint in plane picture itself, can reproduce solid space fully realistically.The process that the plane picture panorama is rebuild stereo-picture comprises: select original image, set up degree of depth tabulation, panorama is rebuild anaglyph, and is three-dimensional synthetic.
Describe each process and the method that realizes that panorama is rebuild below in detail.
The first step is selected original image.Determine exactly to carry out the plane picture that panorama is rebuild, require that its acutance is good, well arranged, sharpness is high, and carried out digitizing.With original image called after T
1, i is listed as the capable pixel value T of j
1(i, j) can with three looks (r, g, b) or four looks (c, m, y, k) expression.
In second step, set up degree of depth tabulation.Degree of depth tabulation is the key that panorama is rebuild, degree of depth tabulation z (i, j) represent pixel point T
1(for convenience tabulation is edited to the degree of depth for i, three-dimensional depth relative value j), we z (i, codomain linear transformation j) is to the integer range of an available byte representation, i.e. integer between 0 to 255, then degree of depth tabulation can convert to and image T
1The 8bit gray level image Z that breadth is identical
1By image T
1Set up gray-scale map Z
1Method have three kinds: the degree of depth is directly filled, the mock-up assisted extraction degree of depth, the three-dimensional model assisted extraction degree of depth.
As a rule, we can be according to image T
1Directly judge the relative depth of picture each several part content, utilize the select tool of Flame Image Process instrument Photoshop software to select regional extent, fill and the corresponding gray-scale value of the degree of depth, set up image Z
1Can make full use of gradual change in the editing environment and fill, duplicate, deepen, subtract instrument such as light, fuzzy and revise local gray-value,, use very conveniently, can obtain stereoeffect preferably though this method precision is low.
The method of the mock-up assisted extraction degree of depth is meant and uses roughly restored image T of the identical mock-up of attitude
1Spatial placement, adopt the whole bag of tricks such as ruler measurements, optical grating projection to obtain the three-dimensional depth of mock-up surface each point, be filled into during the degree of depth tabulates, and set up complete gray level image Z by linear interpolation
1
The method of the three-dimensional model assisted extraction degree of depth is the emphasis that the present invention describes, and this method not only realizes easily, and stereoscopic rendering precision height, effective.In various types of plane pictures, the stereoscopic rendering demand of portrait is the most extensive, realize difficulty the most, so we mainly come the illustration method main points with the process that two-dimensional person's picture is set up degree of depth tabulation, the foundation of other types of image degree of depth tabulation can be with reference to this process realization.
At first, according to image T
1Figure image, as sex, height, fat or thin etc., in three-dimensional animation software, set up three-dimensional model, adjust ratio as parts such as head, neck, trunk, four limbs, and details such as bridge of the nose height, eye socket sinking degree, needn't expend too much energy, if roughly similar to figure image, and need not consider the pinup picture attribute; The second, the attitude of careful adjustment model, and be Real Time Observation in the XY coordinate at front view, up to model attitude and image T
1Personage's attitude try one's best identical till, comprise the stretching, extension situation of limbs, the side appearance angles of facial, health etc. are all consistent, because we need be the Z coordinate figure according to person model three-dimensional depth of visible point in front view, come uncalibrated image T
1In the three-dimensional depth of each point on the portrait, and can not deal with for the part that hair, clothes etc. is difficult to carry out the consistance adjustment, when doing the parallax conversion afterwards, directly revise image Z according to the result of stereoscopy feedback
1Gray-scale value; The 3rd, it is standby to render a width of cloth model attitude image F in front view; The 4th, from three-dimensional animation software, directly read the Z coordinate figure of model each visible point in front view, set up degree of depth tabulation, and linearity changes into gray level image Z
1If, directly do not obtain the function of Z coordinate figure, can adopt the plane intercept method to set up gray level image Z
1, this will further specify below.
Setting up the simplest method of personage's three-dimensional model is the Poser software that adopts Metacreations company to release, and this software provides a large amount of modelling of human body and animal three-dimensional model, can directly select for use, and it is also very convenient to carry out the attitude adjustment in the Poser environment.The 3D Studio MAX groupware of autodesk, inc. exploitation also is the ideal tools of three-dimensional modeling, and being familiar with and understanding it is very useful to the grasp of panorama reconstruction technique.
The plane intercept method can help us to set up gray level image.In three-dimensional coordinate, set up a plane P perpendicular to the Z axle, suitable color is set makes it to have great contrast with model.Plane P can be at Z to moving, and when model and Plane intersects, the part of back is blocked, and can only render the front portion of model.Set initial position, the step-length of plane P, make the plane axially outer mobile along Z, play up M frame animation image in front view, picture number is 1 to M, and M is not more than 255.Create a gray level image Z with the identical size of animated image
1, all gray values of pixel points all are made as 0, animated image is done following processing by numbering order from small to large:
All pixels to m width of cloth animated image are judged, if pixel on plane P, does not deal with, if pixel is on three-dimensional model, with gray-scale value m alternate image Z
1The gray-scale value of last corresponding point, m=1,2,3 ..., M.After all M width of cloth animated images are disposed, can obtain the gray level image Z that a width of cloth is described all pixel three-dimensional depths on the three-dimensional model front view
1, whole process need prepares a computer program and realizes.
For the ease of Flame Image Process, original image T
1, model front view F, gray level image Z
1Size should be consistent with ratio, otherwise need to become identical breadth through picture element interpolation.Model front view F and original image T
1The position of each border or key point is can not be fully one by one corresponding, sometimes even bigger difference, this explanation gray level image Z arranged
1Original image T can't really be described
1Three-dimensional depth, need carry out geometric transformation and adjust image Z
1The position of each pixel.
In view of gray level image Z
1Feature contour not obvious, when geometric transformation, need reference model front view F, set up model front view F and original image T
1Go up the geometric maps relation between each key point, the feature contour, the position of outline line such as personage's eyes, nostril, palm and key point for example, then according to the location map relation with gray level image Z
1Carry out geometric transformation.Geometric transformation just changes the position of pixel, and does not change gray values of pixel points, such as the central point of eyes, the coordinate on image F be (i, j), at image T
1On coordinate be (i ', j '), image Z
1Coordinate be (i, gray-scale value j) has been represented the three-dimensional depth of eye center point, through after the geometric transformation, has moved to position (i ', j '), thereby with image T
1Eye center point really is mapped, therefore, and the gray level image Z after the conversion
1Original image T has accurately been described
1Relative depth.The geometric transformation meeting causes image Z
1Blank spot occurs, need carry out interpolation processing, shareware morpher can finish the geometric maps conversion of gray level image.
In the 3rd step, panorama is rebuild anaglyph.Next just can be according to gray level image Z
1, with image T
1Handle panorama through parallaxial displacement and reconstruct remaining parallax sequence image.Compare image T with stereoscopic photograph
1Be equivalent to first width of cloth in the N width of cloth parallax sequence image, the purpose that panorama is rebuild is according to image T
1Restore other N-1 width of cloth sequence chart.If with image T
1As the intermediate image of parallax sequence chart, the blank spot after the parallax conversion can cause some inconvenience in filling and handling.
Since the existence of parallax, image T
1On pixel T
1(i, j) appear on other disparity map position coordinates for (i+offset, j), offset is the parallaxial displacement amount, with three-dimensional depth z (i, j) and anaglyph numbering relevant:
offset＝[z(i，j)-z
0]*direct*depth*f(z)*k
Wherein, (i j) is pixel T to z
1(i, three-dimensional depth j), z
0Be the three-dimensional depth of parallax free point, that is to say that the degree of depth is z
0The position that on each anaglyph, occurs of pixel be identical, z
0But general value 170; Direct is the direction of parallax conversion, value-1 or+1; Depth is three-dimensional intensive parameter, is set by three-dimensional deviser; F (z) be about parallactic shift amount offset and three-dimensional depth z (i, j) between the correction function of nonlinear relationship,, can not consider when not being very big at three-dimensional depth, directly get f (z)=1; K is the absolute value of rebuilt anaglyph and former anaglyph numbering difference, such as according to the 1st width of cloth image reconstruction n width of cloth image, then k=|1-n|.Every in this manual when relating to three-dimensional depth, we all adopt relative value but not absolute value, and its dirigibility and convenience are conspicuous.
As long as calculate image T according to the parallaxial displacement formula
1Go up each some position in other anaglyph, pixel value is filled into this position, can rebuild whole N width of cloth anaglyphs, the N value can surpass 30 easily, is quite difficult and make and use the stereoscopic camera of 30 camera lenses.In the process of rebuilding anaglyph, can run into three kinds of situations and need special the processing:
(1) image T
1Go up the same position that is in same width of cloth anaglyph after a plurality of different pixel parallaxial displacements, need to judge each point three-dimensional depth z (i, j), select the foremost just the pixel of z value maximum be filled into this position.
(2) since offset value round handle or the pixel transversal displacement after produce some blank spots, mainly appear at the continuum, space of anaglyph, the mild zone of z value variation just need be according to the pixel value linear interpolation filling of peripheral point.
(3) the discontinuous or z value of three-dimensional depth changes the blank spot that more violent zone occurs on anaglyph, and the reason that its corresponding pixel blocks owing to the space is at image T
1Upward invisible or uncorrelated, need three-dimensional deviser rely on the drawing experience manually to fill.
In stereoscopic photograph, every bit imaging in the solid space, the position that occurs in the parallax sequence image is different, has certain transversal displacement, Here it is parallax.Wherein some point can not appear in all anaglyphs owing to the front and back hiding relation, therefore, and in the image of parallax transform reconstruction, a large amount of blank spots can appear, their color is unknown, need fill according to actual conditions, such as top said (2) (3) two kinds of situations.Here the problem that needs to solve is how to guarantee having three-dimensional correlativity between the artificial fill area on all anaglyphs, how reduces the workload of artificial filling.
Effective method is, earlier according to image T
1Rebuild last width of cloth anaglyph T
N, blank spot is carried out interpolation fills and manually fill; Simultaneously, utilize gray level image Z
1Handle self carrying out parallaxial displacement, set up and image T
NCorresponding gray level image Z
N, blank spot is carried out interpolation fill and manually fill, at this moment, the gray-scale value representative image T of filling
NThe three-dimensional depth of corresponding point.Suitably choose the change direction value of direct of parallax, help to simplify the process of artificial filling, improve panorama and rebuild the sense of reality.
At blank map as T
NBlank spot before, can adopt the eye mode, wear the anaglyph spectacles mode and observe stereoscopic image T
1And T
N,, directly revise image Z to the undesirable place of three-dimensional depth
1The gray-scale value of respective regions is rebuild anaglyph T again
N, carrying out stereoscopy, several carries out image Z then till stereoscopic sensation meets the requirements repeatedly
NRebuild and the blank spot filling.Anaglyph T
NNeed artificial zone of filling, its content is at original image T
1In do not exist, no matter fill what color, all do not influence the spatial impression and the completeness of stereo-picture, still, for the outstanding sense of reality, three-dimensional deviser should replenish the clear area according to the content of the part that is blocked as far as possible.
Under the thumping majority situation, stereographic head, last two width of cloth images have comprised all information of intermediate image, that is to say, the every bit on the middle anaglyph is at image T
1Or T
nOn all can find corresponding point.Anaglyph T has been arranged
1, T
NWith gray level image Z
1, Z
N, at this moment N-2 width of cloth sequence image in the middle of we can reconstruct, because information has completeness, does not need middle image is manually filled, and need not set up the gray-scale map of intermediate image correspondence.The concrete grammar of rebuilding intermediate image is:
According to gray level image Z
1, with image T
1Carry out the parallax conversion, changing direction is direct, obtains N-2 width of cloth anaglyph T
2, T
3, T
4..., T
n..., T
N-1
According to gray level image Z
N, with image T
NCarry out the parallax conversion, changing direction is-direct, obtains N-2 width of cloth anaglyph T '
2, T '
3, T '
4..., T '
n..., T '
N-1
Image T
nAnd T '
nBe actually the anaglyph of the identical shooting point that obtains from different conversion approach, the position that their blank spot occurs is different, image T
nOn blank spot, at image T '
nOn can find the color of corresponding point to fill, otherwise too.If also exist the minute quantity blank spot this moment, can manually fill.So far, we have finished the panorama process of reconstruction of all parallax sequence images.
At last, carry out the synthetic of stereo-picture.The parallax sequence image is synthesized raster stereoscopic picture, can adopt optical stereo synthetic, also can adopt digital stereo synthetic, but concrete referenced patent application 981193633 or other article that synthesizes about solid.It should be noted that the value difference that direct gets, the orientation of three-dimensional synthetic middle anaglyph sampling batten is different.
Panorama method for reconstructing provided by the present invention can directly change into stereo-picture very true to nature with plane picture, stereoscopic depth is controlled arbitrarily, stereoscopic sensation is better than the effect that adopts many lens stereos camera to take again, has greatly expanded the application of stereotome.The stereoeffect of panorama method for reconstructing is better than figure layer translation method, and colour of skin texture is better than the stereo-picture that the three-dimensional model pinup picture is played up out, can replace professional stereoscopic camera fully.The solid design that the panorama method for reconstructing can be widely used in professional advertisement more particularly can't realize stereoscopic shooting, or take the too high place of cost, for example international top-brand mouthpiece's flat image directly can be carried out solid and transform.
Provide three case study on implementation below.
Embodiment one, and according to panorama method for reconstructing establishment stereo-picture design software, functional module comprises following content.(1) simple three-dimensional model editing environment.Can create, edit, revise three-dimensional model, and can handle the three-dimensional model that other three-dimensional software such as 3D MAX, POSER set up, can be in view rendering image, can read the three-dimensional coordinate of model surface each point.Utilize this module, can reference plane image creation three-dimensional model, set up degree of depth tabulation, reduced dependence to other three-dimensional software.(2) intercept method three-dimensional depth in plane extracts.Create this module according to the instructions preceding method, can from the animation sequence image that softwares such as 3D MAX are played up by the plane intercept method, directly create gray level image.(3) geometric maps conversion.According to the functional module of the aforementioned main points establishment of instructions, can revise by the gray level image of three-dimensional model foundation and the correspondence position deviation between the plane picture, can reduce the difficulty that three-dimensional model editor and attitude are adjusted.(4) plane picture editing environment.Have that figure layer editor, scope choose, functions such as color is filled, pixel value modification, can directly choose each regional extent of plane picture, fill and the corresponding gray-scale value of three-dimensional depth, set up gray level image, can also finish the artificial filling of anaglyph blank spot.(5) the anaglyph panorama is rebuild.Create this module according to the instructions preceding method, can set that three-dimensional intensive parameter, parallax are changed direction, the width of cloth number of anaglyph, can rebuild whole anaglyphs, and can finish the linear interpolation of blank spot.(6) stereoscopy.Tiling about showing on the screen, the image of stereo format such as tiling or red green (or red indigo plant) glasses up and down adopt modes such as eye, liquid crystal light valve glasses, goggles with color filter are observed stereo-picture, feedback steric information, time update gray level image.(7) three-dimensional synthetic.Extract vertical batten of each anaglyph, rearrange synthetic grating stereo figure.Application possesses the three-dimensional design software of these functional modules, can convert plane picture to stereo-picture very easily.A kind of implicit software processing mode is, calculates after each row of each parallax sequence image according to plane picture and gray-scale map, and it is synthetic directly to carry out grating stereo, and does not export complete parallax sequence image.
Embodiment two, convert personage model's photo to stereo-picture.
Set up personage's three-dimensional model in POSER, the memory model file is input among the 3D MAX, Real Time Observation in front view, and adjustment model attitude and photo are roughly the same.Create a plane vertical with the Z axle, play up 200 frame animation sequence images by the plane intercept method, create gray level image according to these animated images, gray level image is carried out the geometric maps conversion, revise the deviation of correspondence position between gray level image and the plane picture, set then that three-dimensional intensive parameter, parallax are changed direction, the width of cloth number of anaglyph, rebuild whole anaglyphs, and fill in the blanks a little.Extract vertical batten of each anaglyph, rearrange synthetic grating stereo figure, the composite stereo grating.
Embodiment three, convert picture with scenes to stereo-picture.
The front and back hierarchical relationship of picture with scenes is generally apparent in view, and the ground tiling can be chosen ground scope to extending at a distance, the gradual change gray scale is filled wherein, and other scenery depth of field is relatively good to be determined, directly chooses the regional extent of scenery, fill and the corresponding gray-scale value of three-dimensional depth, set up gray level image.According to the transversal displacement amount of each pixel in other anaglyph of gray level image calculating picture with scenes, the filler pixels value is rebuild whole parallax sequence images, carries out solid at last and synthesizes.
Panorama method for reconstructing disclosed by the invention is mainly used in the design of stereo-pictures such as grating three-dimensional picture, laser hologram picture, three-dimensional film, stereotelevision, and another important application is an exploitation stereo-picture design software.Aim of the present invention is, set up the degree of depth tabulation of all pixels according to the plane picture content, utilize degree of depth list value to rebuild other parallax sequence image, the blank spot of disappearance information is filled required color, understand the rudimentary knowledge that this instructions needs to be grasped plane picture processing and three-dimensional modeling.The present invention has chosen three case study on implementation and has come with reference to explanation, and the solution that the professional person provides according to this instructions can be easy to write out a lot of relevant case study on implementation, and intention of the present invention will comprise all embodiment that meet with claims.
Claims (8)
1, a kind of panorama method for reconstructing, one width of cloth plane picture directly can be changed into the parallax sequence image that can constitute stereo-picture, it is characterized in that, according to the space distribution of plane picture, set up the tabulation that comprises each pixel three-dimensional depth, utilize degree of depth tabulation to calculate the transversal displacement amount of each pixel in other anaglyph of plane picture, pixel value is filled in these anaglyphs, reconstructs all parallax sequence images.
2, the method for claim 1, its feature also is, be in the same position of same width of cloth anaglyph behind a plurality of different pixel transversal displacements of plane picture, (i j), selects the maximum pixel of z value to be filled into this position to the three-dimensional depth z of each pixel of needs judgement.
3, method as claimed in claim 2, its feature also are, appear at the blank spot of the continuum, space of anaglyph, fill according to the pixel value linear interpolation of peripheral point; Appear at the discontinuous or violent regional blank spot of z value variation of anaglyph three-dimensional depth, manually fill.
4, method as claimed in claim 3, its feature are that also the order of rebuilding anaglyph is, earlier according to image T
1With gray-scale map Z
1Rebuild last width of cloth anaglyph T
NWith gray level image Z
N, to T
N, Z
NBlank spot is filled, then to image sets T
1, Z
1And T
N, Z
NCarry out parallaxial displacement respectively and handle, reconstruct two groups of parallax sequence images, the blank spot that occurs on the anaglyph can be organized at another and find corresponding point to fill on identical anaglyph of sequence number.
5, the method for claim 1, its feature also be, the integer of the codomain linear transformation to 0 of degree of depth tabulation between 255 converts degree of depth tabulation to the 8bit gray level image.
6, the method for claim 1, its feature also are, directly choose each regional extent of plane picture, fill and the corresponding gray-scale value of three-dimensional depth, set up gray level image.
7, the method for claim 1, its feature also is, set up three-dimensional model according to plane picture, adjust the attitude of three-dimensional model, set up the degree of depth tabulation of front view, and change into gray level image, and gray level image is carried out the geometric maps conversion, revise the position of gray level image pixel.
8, method as claimed in claim 7, its feature are that also gray level image adopts three-dimensional model plane intercept method to set up.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB031376606A CN100414566C (en) | 2003-06-19 | 2003-06-19 | Panoramic reconstruction method of three dimensional image from two dimensional image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB031376606A CN100414566C (en) | 2003-06-19 | 2003-06-19 | Panoramic reconstruction method of three dimensional image from two dimensional image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1567385A true CN1567385A (en) | 2005-01-19 |
| CN100414566C CN100414566C (en) | 2008-08-27 |
Family
ID=34470491
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB031376606A Expired - Fee Related CN100414566C (en) | 2003-06-19 | 2003-06-19 | Panoramic reconstruction method of three dimensional image from two dimensional image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100414566C (en) |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100382567C (en) * | 2006-06-06 | 2008-04-16 | 王程 | Method for rebuilding super resolution image from reduced quality image caused by interlaced sampling |
| CN101562755B (en) * | 2009-05-19 | 2010-09-01 | 无锡景象数字技术有限公司 | Method for producing 3D video by plane video |
| CN101908233A (en) * | 2010-08-16 | 2010-12-08 | 福建华映显示科技有限公司 | Method and system for producing plural viewpoint picture for three-dimensional image reconstruction |
| CN101917643A (en) * | 2010-07-09 | 2010-12-15 | 清华大学 | Method and device for detecting lens in real time in fully automatic two-dimensional (2D) to three-dimensional (3D) technology |
| CN101563930B (en) * | 2006-12-28 | 2011-05-25 | 日本电信电话株式会社 | Video encoding method, decoding method, device thereof, program thereof, and storage medium containing the program |
| CN102216979A (en) * | 2009-01-21 | 2011-10-12 | 索尼公司 | Image processing apparatus, image processing method and program |
| CN102223518A (en) * | 2010-04-14 | 2011-10-19 | 鸿富锦精密工业（深圳）有限公司 | Stereoscopic digital imaging monitoring system and stereoscopic digital imaging monitoring method |
| CN101646022B (en) * | 2009-09-04 | 2011-11-16 | 华为终端有限公司 | Image splicing method and system thereof |
| CN101841727B (en) * | 2009-12-30 | 2011-12-07 | 福州华映视讯有限公司 | Image processing method |
| CN101754042B (en) * | 2008-10-30 | 2012-07-11 | 华为终端有限公司 | Image reconstruction method and image reconstruction system |
| CN101656835B (en) * | 2008-08-21 | 2012-10-31 | 索尼株式会社 | Image pickup apparatus, display and image processing apparatus |
| CN102903143A (en) * | 2011-07-27 | 2013-01-30 | 国际商业机器公司 | Method and system for converting two-dimensional image into three-dimensional image |
| CN103024416A (en) * | 2011-12-14 | 2013-04-03 | 微软公司 | Parallax compensation |
| CN104899921A (en) * | 2015-06-04 | 2015-09-09 | 杭州电子科技大学 | Single-view video human body posture recovery method based on multi-mode self-coding model |
| CN105374067A (en) * | 2015-10-10 | 2016-03-02 | 长安大学 | Three-dimensional reconstruction method based on PAL cameras and reconstruction system thereof |
| US9324184B2 (en) | 2011-12-14 | 2016-04-26 | Microsoft Technology Licensing, Llc | Image three-dimensional (3D) modeling |
| US9406153B2 (en) | 2011-12-14 | 2016-08-02 | Microsoft Technology Licensing, Llc | Point of interest (POI) data positioning in image |
| CN106127862A (en) * | 2016-06-16 | 2016-11-16 | 财付通支付科技有限公司 | The treating method and apparatus of figure |
| CN107292963A (en) * | 2016-04-12 | 2017-10-24 | 杭州海康威视数字技术股份有限公司 | The method of adjustment and device of a kind of threedimensional model |
| CN108064447A (en) * | 2017-11-29 | 2018-05-22 | 深圳前海达闼云端智能科技有限公司 | Method for displaying image, intelligent glasses and storage medium |
| US10038842B2 (en) | 2011-11-01 | 2018-07-31 | Microsoft Technology Licensing, Llc | Planar panorama imagery generation |
| CN110197529A (en) * | 2018-08-30 | 2019-09-03 | 杭州维聚科技有限公司 | Interior space three-dimensional rebuilding method |
| CN110544299A (en) * | 2019-09-05 | 2019-12-06 | 广东电网有限责任公司 | power distribution network 10kV pole load switch 3D model and generation method thereof |
| CN113804183A (en) * | 2021-09-17 | 2021-12-17 | 广东汇天航空航天科技有限公司 | Real-time topographic surveying and mapping method and system |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100445209B1 (en) * | 1995-12-19 | 2004-12-13 | 코닌클리케 필립스 일렉트로닉스 엔.브이. | Image processing system and image conversion processor for generating input images into at least one output image through parallax conversion |
| US6031564A (en) * | 1997-07-07 | 2000-02-29 | Reveo, Inc. | Method and apparatus for monoscopic to stereoscopic image conversion |
| WO1999030280A1 (en) * | 1997-12-05 | 1999-06-17 | Dynamic Digital Depth Research Pty. Ltd. | Improved image conversion and encoding techniques |
| CA2418089A1 (en) * | 2000-08-04 | 2002-02-14 | Dynamic Digital Depth Research Pty Ltd. | Image conversion and encoding technique |
| CN1154073C (en) * | 2001-02-15 | 2004-06-16 | 矽统科技股份有限公司 | 3D space image converter with regulatable stereoscopic effect and its method |
-
2003
- 2003-06-19 CN CNB031376606A patent/CN100414566C/en not_active Expired - Fee Related
Cited By (34)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100382567C (en) * | 2006-06-06 | 2008-04-16 | 王程 | Method for rebuilding super resolution image from reduced quality image caused by interlaced sampling |
| CN101563930B (en) * | 2006-12-28 | 2011-05-25 | 日本电信电话株式会社 | Video encoding method, decoding method, device thereof, program thereof, and storage medium containing the program |
| CN101656835B (en) * | 2008-08-21 | 2012-10-31 | 索尼株式会社 | Image pickup apparatus, display and image processing apparatus |
| CN101754042B (en) * | 2008-10-30 | 2012-07-11 | 华为终端有限公司 | Image reconstruction method and image reconstruction system |
| CN102216979A (en) * | 2009-01-21 | 2011-10-12 | 索尼公司 | Image processing apparatus, image processing method and program |
| CN101562755B (en) * | 2009-05-19 | 2010-09-01 | 无锡景象数字技术有限公司 | Method for producing 3D video by plane video |
| CN101646022B (en) * | 2009-09-04 | 2011-11-16 | 华为终端有限公司 | Image splicing method and system thereof |
| CN101841727B (en) * | 2009-12-30 | 2011-12-07 | 福州华映视讯有限公司 | Image processing method |
| CN102223518A (en) * | 2010-04-14 | 2011-10-19 | 鸿富锦精密工业（深圳）有限公司 | Stereoscopic digital imaging monitoring system and stereoscopic digital imaging monitoring method |
| CN101917643A (en) * | 2010-07-09 | 2010-12-15 | 清华大学 | Method and device for detecting lens in real time in fully automatic two-dimensional (2D) to three-dimensional (3D) technology |
| CN101917643B (en) * | 2010-07-09 | 2012-05-30 | 清华大学 | Method and device for detecting lens in real time in fully automatic two-dimensional (2D) to three-dimensional (3D) technology |
| CN101908233A (en) * | 2010-08-16 | 2010-12-08 | 福建华映显示科技有限公司 | Method and system for producing plural viewpoint picture for three-dimensional image reconstruction |
| CN102903143A (en) * | 2011-07-27 | 2013-01-30 | 国际商业机器公司 | Method and system for converting two-dimensional image into three-dimensional image |
| US10038842B2 (en) | 2011-11-01 | 2018-07-31 | Microsoft Technology Licensing, Llc | Planar panorama imagery generation |
| US10008021B2 (en) | 2011-12-14 | 2018-06-26 | Microsoft Technology Licensing, Llc | Parallax compensation |
| US9324184B2 (en) | 2011-12-14 | 2016-04-26 | Microsoft Technology Licensing, Llc | Image three-dimensional (3D) modeling |
| US9406153B2 (en) | 2011-12-14 | 2016-08-02 | Microsoft Technology Licensing, Llc | Point of interest (POI) data positioning in image |
| CN103024416B (en) * | 2011-12-14 | 2015-06-17 | 微软公司 | Parallax compensation |
| CN103024416A (en) * | 2011-12-14 | 2013-04-03 | 微软公司 | Parallax compensation |
| CN104899921A (en) * | 2015-06-04 | 2015-09-09 | 杭州电子科技大学 | Single-view video human body posture recovery method based on multi-mode self-coding model |
| CN104899921B (en) * | 2015-06-04 | 2017-12-22 | 杭州电子科技大学 | Single-view videos human body attitude restoration methods based on multi-modal own coding model |
| CN105374067A (en) * | 2015-10-10 | 2016-03-02 | 长安大学 | Three-dimensional reconstruction method based on PAL cameras and reconstruction system thereof |
| US10665016B2 (en) | 2016-04-12 | 2020-05-26 | Hangzhou Hikvision Digital Technology Co., Ltd. | Method and device for adjusting three-dimensional model |
| CN107292963A (en) * | 2016-04-12 | 2017-10-24 | 杭州海康威视数字技术股份有限公司 | The method of adjustment and device of a kind of threedimensional model |
| CN107292963B (en) * | 2016-04-12 | 2020-01-17 | 杭州海康威视数字技术股份有限公司 | Three-dimensional model adjusting method and device |
| CN106127862B (en) * | 2016-06-16 | 2019-02-26 | 财付通支付科技有限公司 | The treating method and apparatus of figure |
| CN106127862A (en) * | 2016-06-16 | 2016-11-16 | 财付通支付科技有限公司 | The treating method and apparatus of figure |
| CN108064447A (en) * | 2017-11-29 | 2018-05-22 | 深圳前海达闼云端智能科技有限公司 | Method for displaying image, intelligent glasses and storage medium |
| CN110197529A (en) * | 2018-08-30 | 2019-09-03 | 杭州维聚科技有限公司 | Interior space three-dimensional rebuilding method |
| CN110197529B (en) * | 2018-08-30 | 2022-11-11 | 杭州维聚科技有限公司 | Indoor space three-dimensional reconstruction method |
| CN110544299A (en) * | 2019-09-05 | 2019-12-06 | 广东电网有限责任公司 | power distribution network 10kV pole load switch 3D model and generation method thereof |
| CN110544299B (en) * | 2019-09-05 | 2020-10-16 | 广东电网有限责任公司 | Power distribution network 10kV pole load switch 3D model and generation method thereof |
| CN113804183A (en) * | 2021-09-17 | 2021-12-17 | 广东汇天航空航天科技有限公司 | Real-time topographic surveying and mapping method and system |
| CN113804183B (en) * | 2021-09-17 | 2023-12-22 | 广东汇天航空航天科技有限公司 | Real-time topographic mapping method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100414566C (en) | 2008-08-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100414566C (en) | Panoramic reconstruction method of three dimensional image from two dimensional image | |
| US7643025B2 (en) | Method and apparatus for applying stereoscopic imagery to three-dimensionally defined substrates | |
| US9288476B2 (en) | System and method for real-time depth modification of stereo images of a virtual reality environment | |
| US20120182403A1 (en) | Stereoscopic imaging | |
| CN102568026B (en) | Three-dimensional enhancing realizing method for multi-viewpoint free stereo display | |
| CN103974055B (en) | 3D photo generation system and method | |
| KR100327541B1 (en) | 3D facial modeling system and modeling method | |
| US20060152579A1 (en) | Stereoscopic imaging system | |
| CN1641702A (en) | Method for designing stereo image from planar image | |
| CN102572482A (en) | 3D (three-dimensional) reconstruction method for stereo/multi-view videos based on FPGA (field programmable gata array) | |
| CN102819855B (en) | The generation method of two dimensional image and device | |
| Taylor et al. | Reconstructing polyhedral models of architectural scenes from photographs | |
| CN104091366B (en) | Three-dimensional intelligent digitalization generation method and system based on two-dimensional shadow information | |
| AU2004306226B2 (en) | Stereoscopic imaging | |
| Verhoeven | Computer graphics meets image fusion: The power of texture baking to simultaneously visualise 3D surface features and colour | |
| CN116152417B (en) | Multi-viewpoint perspective space fitting and rendering method and device | |
| Gatesy et al. | Anaglyph stereo imaging of dinosaur track morphology and microtopography | |
| CN102467747A (en) | Building decoration animation three-dimensional (3D) effect processing method | |
| CN116310150B (en) | Furniture multi-view three-dimensional model reconstruction method based on multi-scale feature fusion | |
| CN118429595B (en) | Motion blur implementation method based on three-dimensional engine | |
| Feldman et al. | Interactive 2D to 3D stereoscopic image synthesis | |
| WO2017031117A1 (en) | System and method for real-time depth modification of stereo images of a virtual reality environment | |
| Yun et al. | Development of Stereoscopic Image Editing Tool using Image-Based Modeling. | |
| Rizvic et al. | Maya Sun Simulation of Bosnian Gravestone Virtual Model. | |
| Wang et al. | Review of multi view auto-stereoscopic display system based on depth image analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20080827 Termination date: 20130619 |