CN109118576A - Large scene three-dimensional reconstruction system and method for reconstructing based on BDS location-based service - Google Patents
Large scene three-dimensional reconstruction system and method for reconstructing based on BDS location-based service Download PDFInfo
- Publication number
- CN109118576A CN109118576A CN201810794467.6A CN201810794467A CN109118576A CN 109118576 A CN109118576 A CN 109118576A CN 201810794467 A CN201810794467 A CN 201810794467A CN 109118576 A CN109118576 A CN 109118576A
- Authority
- CN
- China
- Prior art keywords
- texture
- space
- image
- mapping
- scene
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T15/00—3D [Three Dimensional] image rendering
- G06T15/04—Texture mapping
-
- 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
- G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2219/00—Indexing scheme for manipulating 3D models or images for computer graphics
- G06T2219/20—Indexing scheme for editing of 3D models
- G06T2219/2012—Colour editing, changing, or manipulating; Use of colour codes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2219/00—Indexing scheme for manipulating 3D models or images for computer graphics
- G06T2219/20—Indexing scheme for editing of 3D models
- G06T2219/2024—Style variation
Abstract
The invention discloses a kind of method for reconstructing of large scene three-dimensional reconstruction system based on BDS location-based service, it is characterized in that, described method includes following steps: S1. obtains scene image to progress scene image positioning by BDS location-based service module and establishes panoramic picture image collection system；S2. it carries out data processing and is stored in database；Model construction under the space S3.Grassmann；S4. the mapping of texture and color.This method can obtain scene image in real time, precisely match, can rapidly carry out three-dimensional modeling, and can overcome the problems, such as that accumulated error is excessive in traditional large scene three-dimensional reconstruction, accessing cost for data space is huge.
Description
Technical field
The present invention relates to scene three-dimensional reconstruction fields, more particularly to the large scene based on BDS location-based service is three-dimensional
Reconstructing system and method for reconstructing.
Background technique
The three-dimensional reconstruction of large scale scene has become one of the development trend of future computer vision, and threedimensional model
The 4th kind of multimedia data type after image, sound and video is had become, and in such as mapping, urban planning, public security
The fields such as criminal investigation also have important, broad application prospect.Current existing large scene three-dimensional rebuilding method is all often base
In the image rebuilding method of common camera model, limitation and image reconstruction due to common computer visual angle are not suitable for greatly
The three-dimensional reconstruction of type complex scene, therefore, there is an urgent need to propose a kind of large scene three-dimensional rebuilding method of highly effective.
Realize that the method for three-dimensional modeling substantially has following a few classes at present: one kind is directly such as to be counted using 3 d modeling software
Calculation machine Autocad (AutoCAD), animation rendering Software for producing (3D Max, Maya etc.) etc.；This tool-type it is man-machine
Interactive modeling can construct complicated geometric scene using basic geometric element, and application technology is skillful, and method is flexible, and energy is true to nature again
The geometry and surface texture information of existing object, however the characteristics of due to its fine modeling, take so that modeling data amount is huge
When it is laborious, and there is certain requirement to computer hardware, thus be not suitable for the three-dimensional modeling of large scene；
Another kind of is that the data information based on photogrammetric three-dimensional modeling, by acquiring real-world object constructs model.It adopts
Collect the general sampling depth information of data and camera captures image, obtains threedimensional model after registration, fusion, integration.But
The reconstruction that limitation and image reconstruction due to computer visual angle are not suitable for large complicated scene thus becomes large-scale three dimensional scene
The bottleneck of reconstruction；
It is the three-dimensional modeling based on three-dimensional point cloud scanning technique there are also one kind.With data acquisition technology continuous development and
Automation, three-dimensional laser point cloud data constructs the hot spot that threedimensional model has become three-dimensional modeling automatically, to building point cloud data
The recurrence step of model specifically includes that the boundary characteristic on the one hand extracting building, is constraint building 3D solid mould with feature
Type；On the other hand topological relation directly is established to meshing point cloud data, carries out resurfacing and optimization.Three based on cloud
Dimension modeling overcomes the deficiency of traditional data acquisition mode, improves the precision and efficiency of data acquisition.But since three-dimensional is swept
Retouch to obtain in point cloud data and be easy to produce loophole and point cloud data amount is very huge, also constrain its in large-scale modeling into one
Step develops and is widely applied.
Summary of the invention
The purpose of the present invention is in view of the deficiencies of the prior art, and provide a kind of large scene based on BDS location-based service
Three-dimensional reconstruction system and method for reconstructing.This system realizes quick convenience, data processing, three-dimensional reconstruction efficiently and accurately and cost
It is lower.This method can obtain scene image in real time, precisely match, can rapidly carry out three-dimensional modeling, and can overcome
The problem that accumulated error is excessive in traditional large scene three-dimensional reconstruction, accessing cost for data space is huge.
Realizing the technical solution of the object of the invention is:
Large scene three-dimensional reconstruction system based on BDS location-based service, unlike the prior art, the system comprises
BDS location-based service module, data processing module, three-dimensional reconstruction module and the texture mapping module of sequential connection.
The BDS location-based service module, which is equipped with, to be supported the receiving antenna unit of active antenna and passive antenna, is used for LNA days
Line detection radio-frequency front-end integrated unit and for the position signal of detection and the image data of acquisition to be sent to service background
Communication unit.
The data processing module is that the image obtained to server background carries out pretreatment and feature extraction, builds number
According to library, by treated, data are stored in database, and the pretreatment of image includes image enhancement and image denoising, feature extraction packet
The pixel for extracting image, color value and spatial relation feature are included, database is built according to the image based on content
The architecture of retrieval (Content Based Image Retrieval, abbreviation CBIR) is designed.
The three-dimensional reconstruction module, the moulding modeling including deepness image registration and three-dimension object curve and surface, wherein deep
Image registration is spent, is integrated under complete consistent coordinate system including the three-dimensional data points to multi-directional scanning, three-dimension object curve is bent
The moulding in face models, and is the particle being mapped as the pixel in image space under the space Grassmann, using
The reasonable Recurrent curve algorithm confrontation point redefined under the space Grassmann carries out three-dimensional reconstruction.
The texture mapping module, including texture mapping and texture load, wherein texture object relies on texture space and figure
The mapping mode of image space, texture mapping are divided into forward texture mapping and inverse texture, and the support of texture loading unit is based on
The texture of OpenGL loads.
With the method for reconstructing of the above-mentioned large scene three-dimensional reconstruction system based on BDS location-based service, include the following steps:
S1. scene image positioning is carried out by BDS location-based service module, obtain scene image first and establishes panoramic picture
Acquisition system: the scene image includes the scene picture that common camera obtains, and the panorama sketch acquisition system includes to three
The target identification of dimensional object and tracking shoot and obtain scene color image and depth image to each orientation, are relied on
BDS location-based service real-time dynamicly detects surrounding scene changes, time service precision 20ns, and positioning accuracy 10m needs position to take
When business, location data is sent to service background；
S2. it carries out data processing and establishes scene image data library: the obtained image of step S1 is enhanced, is denoised,
The processing such as segmentation, carries out feature extraction, extracts the pixel, color and positional relationship feature of image, extracted feature is for spy
Sign compares and the similarity measurement of characteristic vector, and by treated, data are stored in scene image data library；
Model construction under the space S3.Grassmann: the data point that step S2 is obtained is converted into the space Grassmann
Under particle；Reasonable Recurrent curve curved surface algorithm is redefined under the space Grassmann simultaneously, under the space Grassmann
Recurrent curve Applications of Surface Modeling Methods carry out three-dimensional modeling, under the space Grassmann the definition of Recurrent curve be with reasonable L-curve
Example, reasonable L-curve are as follows in the recursive expression of affine space:
(mP, m) and (v, 0) is the basic element in the space Grassmann, and P and v have affine space, in addition, weight
The particle for being zero is considered as the vector in the space Grassmann, therefore, under the space Grassmann, it is bent to redefine reasonable L
The standard literary style of line is as follows:
It is worth noting that, reasonable L-curve is made of the point of affine space, but its control structure (ωiPi,ωi) be present in
The space Grassmann；
S4. the mapping of texture and color: the mapping of color and texture is carried out to the threedimensional model that step S3 is rebuild, is restored
Sense of reality large scene, wherein the mapping of texture and the color is based on underlying graphics library (Open Graphics
Library, abbreviation OpenGL) and visualization tools function storehouse (visualization toolkit, abbreviation vtk) and Qt open
Send out frame, process are as follows:
1) color mapping and Texture Matching are carried out to tri patch based on shape library OpenGL；
2) define threedimensional model texture: sampling function discrete method is defined texture, according to domain difference, function line
Reason is divided into one-dimensional texture, 2 d texture and three-D grain；
3) it specifies mapping mode: mapping mode being obtained according to texture space and image space, texture mapping is divided into forward texture
Mapping and inverse texture；
4) color and texture are loaded: based on OpenGL shape library load color and texture.
This system realizes quick convenience, data processing, three-dimensional reconstruction efficiently and accurately and cost is relatively low.This method can
It to obtain scene image in real time, precisely matches, can rapidly carry out three-dimensional modeling, and traditional large scene can be overcome three-dimensional
The problem that accumulated error is excessive in reconstruction, accessing cost for data space is huge.
Detailed description of the invention
Fig. 1 is the system structure diagram of embodiment；
Fig. 2 is the method flow schematic diagram of embodiment.
Specific embodiment
The content of present invention is further elaborated with reference to the accompanying drawings and examples, but is not limitation of the invention.
Embodiment:
Referring to Fig.1, based on the large scene three-dimensional reconstruction system of BDS location-based service, the system comprises sequential connections
BDS location-based service module, data processing module, three-dimensional reconstruction module and texture mapping module.
The BDS location-based service module, which is equipped with, to be supported the receiving antenna unit of active antenna and passive antenna, is used for LNA days
Line detection radio-frequency front-end integrated unit and for the position signal of detection and the image data of acquisition to be sent to service background
Communication unit, this example is using the ATGM336H-5N- in BDS/GNSS full constellation positioning navigation module ATGM336H-5N series
2X module.
The data processing module, the data including obtaining to server background carry out pretreatment and feature extraction, finally
It writes data into database, the pretreatment of image includes image enhancement and image denoising etc., and feature extraction includes extracting figure
Pixel, color value and the spatial relation feature of picture, database is according to content-based image retrieval (Content
Based Image Retrieval, abbreviation CBIR) architecture be designed.
The three-dimensional reconstruction module, the moulding modeling including deepness image registration and three-dimension object curve and surface, wherein deep
Image registration is spent, is integrated under complete consistent coordinate system including the three-dimensional data points to multi-directional scanning, three-dimension object curve is bent
The moulding in face models, and is the particle being mapped as the pixel in image space under the space Grassmann, using
The reasonable Recurrent curve algorithm confrontation point redefined under the space Grassmann carries out three-dimensional reconstruction.
The texture mapping module, including texture mapping and texture load, wherein texture object relies on texture space and figure
The mapping mode of image space, texture mapping are divided into forward texture mapping and inverse texture, and the support of texture loading unit is based on
The texture of OpenGL loads.
Referring to Fig. 2, with the method for reconstructing of the above-mentioned large scene three-dimensional reconstruction system based on BDS location-based service, including such as
Lower step:
S1. by BDS location-based service module photograph scene image go forward side by side line trace positioning, the BDS location-based service relied on
Surrounding scene changes are real-time dynamicly detected, time service precision 20ns, positioning accuracy 10m when needing location-based service, will be positioned
Data are sent to service background；
S2. data processing and deposit database are carried out: the obtained image of step S1 is enhanced, denoises, divides etc.
Reason carries out feature extraction, extracts the pixel, color and positional relationship feature of image, extracted feature for aspect ratio to
And the similarity measurement of characteristic vector, by treated, data are stored in scene image data library；
Model construction under the space S3.Grassmann: the data point that step S2 is obtained is converted into the space Grassmann
Under particle；Reasonable Recurrent curve curved surface algorithm is redefined under the space Grassmann simultaneously, under the space Grassmann
Recurrent curve Applications of Surface Modeling Methods carry out three-dimensional modeling, typically, reasonable L-curve affine space recursive expression such as
Under:
(mP, m) and (v, 0) is the basic element in the space Grassmann, and P and v have affine space, in addition, weight
The particle for being zero is considered as the vector in the space Grassmann, therefore, under the space Grassmann, it is bent to redefine reasonable L
The standard literary style of line is as follows:
It is worth noting that, reasonable L-curve is made of the point of affine space, but its control structure (ωiPi,ωi) be present in
The space Grassmann；
S4. the mapping of texture and color: the mapping of color and texture is carried out to the threedimensional model that step S3 is rebuild, is restored
Sense of reality large scene, wherein the mapping of texture and the color is based on underlying graphics library (Open Graphics
Library, abbreviation OpenGL) and visualization tools function storehouse (visualization toolkit, abbreviation vtk) and Qt open
Send out frame, process are as follows:
1) color mapping and Texture Matching are carried out to tri patch based on shape library OpenGL；
2) define threedimensional model texture: sampling function discrete method is defined texture, according to domain difference, function line
Reason is divided into one-dimensional texture, 2 d texture and three-D grain；
3) it specifies mapping mode: mapping mode being obtained according to texture space and image space, texture mapping is divided into forward texture
Mapping and inverse texture；
4) color and texture are loaded: based on OpenGL shape library load color and texture.
Claims (6)
1. the large scene three-dimensional reconstruction system based on BDS location-based service, which is characterized in that the system comprises sequential connections
BDS location-based service module, data processing module, three-dimensional reconstruction module and texture mapping module.
2. as described in claim 1 based on the large scene three-dimensional reconstruction system of BDS location-based service, which is characterized in that described
BDS location-based service module is equipped with the communication unit of receiving antenna unit, radio-frequency front-end integrated unit sum.
3. as described in claim 1 based on the large scene three-dimensional reconstruction system of BDS location-based service, which is characterized in that described
Data processing module is that the image data obtained to server background carries out pretreatment and feature extraction, will treated data
It is stored in database, the pretreatment of image includes image enhancement and image denoising, and feature extraction includes extracting the pixel of image
Point, color value and spatial relation feature, database are designed according to the architecture of content-based image retrieval.
4. as described in claim 1 based on the large scene three-dimensional reconstruction system of BDS location-based service, it is characterised in that three dimensional field
Scape modeling, is the particle under the space Grassmann by the Mapping of data points in image space, using under the space Grassmann
The reasonable Recurrent curve algorithm confrontation point redefined carries out three-dimensional reconstruction.
5. as described in claim 1 based on the large scene three-dimensional reconstruction system of BDS location-based service, which is characterized in that described
Texture mapping module includes the definition and texture load of texture object, wherein texture object support texture space and image space
Mapping mode, texture mapping are divided into forward texture mapping and inverse texture, and texture loading unit is supported based on OpenGL's
Texture load.
6. using the reconstruction of the large scene three-dimensional reconstruction system based on BDS location-based service described in claim 1-5 any one
Method, which is characterized in that described method includes following steps:
S1. scene image is obtained to progress scene image positioning by BDS location-based service module and establishes panoramic picture acquisition system
System: the scene image includes the scene picture that common camera obtains, and the panorama sketch acquisition system includes to three dimensional object
Target identification and tracking be each orientation shoot and obtain scene color image and depth image, the position BDS relied on
Service real-time dynamicly detects surrounding scene changes, and time service precision 20ns, positioning accuracy 10m will when needing location-based service
Location data is sent to service background；
S2. it carries out data processing and is stored in database: the obtained image of step S1 being enhanced, is denoised, dividing processing, is carried out
Feature extraction, extracts the pixel, color and positional relationship feature of image, and extracted feature is used for aspect ratio pair and Characteristic Vectors
The similarity measurement of amount, by treated, data are stored in database；
Model construction under the space S3.Grassmann: the data point that step S2 is obtained is converted under the space Grassmann
Particle；Reasonable Recurrent curve curved surface algorithm is redefined under the space Grassmann simultaneously, with passing under the space Grassmann
Curve and surface modeling method is returned to carry out three-dimensional modeling, reasonable L-curve is as follows in the recursive expression of affine space:
(mP, m) and (v, 0) is the basic element in the space Grassmann, and P and v have affine space, in addition, weight is zero
Particle be considered as therefore vector in the space Grassmann under the space Grassmann, redefines reasonable L-curve
Standard literary style is as follows:
Reasonable L-curve is made of the point of affine space, but its control structure (ωiPi,ωi) it is present in the space Grassmann；
S4. the mapping of texture and color: carrying out the mapping of color and texture to the threedimensional model that step S3 is rebuild, and restores true
Large scene is felt, wherein the mapping of texture and the color is based on underlying graphics library (Open Graphics Library, letter
Claim OpenGL) and visualization tools function storehouse (visualization toolkit, abbreviation vtk) and Qt Development Framework, process
Are as follows:
1) color mapping and Texture Matching are carried out to tri patch based on shape library OpenGL；
2) define threedimensional model texture: sampling function discrete method is defined texture, according to domain difference, function texture point
For one-dimensional texture, 2 d texture and three-D grain；
3) it specifies mapping mode: mapping mode being obtained according to texture space and image space, texture mapping is divided into forward texture mapping
And inverse texture；
4) color and texture are loaded: based on OpenGL shape library load color and texture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810794467.6A CN109118576A (en) | 2018-07-19 | 2018-07-19 | Large scene three-dimensional reconstruction system and method for reconstructing based on BDS location-based service |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810794467.6A CN109118576A (en) | 2018-07-19 | 2018-07-19 | Large scene three-dimensional reconstruction system and method for reconstructing based on BDS location-based service |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109118576A true CN109118576A (en) | 2019-01-01 |
Family
ID=64862946
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810794467.6A Pending CN109118576A (en) | 2018-07-19 | 2018-07-19 | Large scene three-dimensional reconstruction system and method for reconstructing based on BDS location-based service |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109118576A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111768487A (en) * | 2020-06-11 | 2020-10-13 | 武汉市工程科学技术研究院 | Geological rock data three-dimensional reconstruction system and method based on three-dimensional point cloud base |
| CN112489064A (en) * | 2020-12-14 | 2021-03-12 | 桂林电子科技大学 | Panorama segmentation method based on edge scaling correction |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2820269A1 (en) * | 2001-01-30 | 2002-08-02 | Koninkl Philips Electronics Nv | PROCESS FOR PROCESSING 2D IMAGES APPLIED TO 3D OBJECTS |
| CN102724287A (en) * | 2012-05-18 | 2012-10-10 | 南京大学 | Mobile space four dimensional information service system and terminals thereof and location method |
| US8339394B1 (en) * | 2011-08-12 | 2012-12-25 | Google Inc. | Automatic method for photo texturing geolocated 3-D models from geolocated imagery |
| CN103021017A (en) * | 2012-12-04 | 2013-04-03 | 上海交通大学 | Three-dimensional scene rebuilding method based on GPU acceleration |
| CN104021586A (en) * | 2014-05-05 | 2014-09-03 | 深圳市城市管理监督指挥中心 | Air-ground integrated city ecological civilization managing system and method based on Beidou positioning |
| CN204359937U (en) * | 2015-01-23 | 2015-05-27 | 桂林电子科技大学 | A kind of based on big-dipper satellite location and the GPS simulation system of time service |
| US20160119541A1 (en) * | 2014-10-24 | 2016-04-28 | Bounce Imaging, Inc. | Imaging systems and methods |
| CN107491853A (en) * | 2017-06-22 | 2017-12-19 | 杨雪榕 | Parallel computing pilot system and test method based on emulation in loop |
| CN107622525A (en) * | 2017-10-09 | 2018-01-23 | 贵州信鸽科技有限公司 | Threedimensional model preparation method, apparatus and system |
-
2018
- 2018-07-19 CN CN201810794467.6A patent/CN109118576A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2820269A1 (en) * | 2001-01-30 | 2002-08-02 | Koninkl Philips Electronics Nv | PROCESS FOR PROCESSING 2D IMAGES APPLIED TO 3D OBJECTS |
| US8339394B1 (en) * | 2011-08-12 | 2012-12-25 | Google Inc. | Automatic method for photo texturing geolocated 3-D models from geolocated imagery |
| CN102724287A (en) * | 2012-05-18 | 2012-10-10 | 南京大学 | Mobile space four dimensional information service system and terminals thereof and location method |
| CN103021017A (en) * | 2012-12-04 | 2013-04-03 | 上海交通大学 | Three-dimensional scene rebuilding method based on GPU acceleration |
| CN104021586A (en) * | 2014-05-05 | 2014-09-03 | 深圳市城市管理监督指挥中心 | Air-ground integrated city ecological civilization managing system and method based on Beidou positioning |
| US20160119541A1 (en) * | 2014-10-24 | 2016-04-28 | Bounce Imaging, Inc. | Imaging systems and methods |
| CN204359937U (en) * | 2015-01-23 | 2015-05-27 | 桂林电子科技大学 | A kind of based on big-dipper satellite location and the GPS simulation system of time service |
| CN107491853A (en) * | 2017-06-22 | 2017-12-19 | 杨雪榕 | Parallel computing pilot system and test method based on emulation in loop |
| CN107622525A (en) * | 2017-10-09 | 2018-01-23 | 贵州信鸽科技有限公司 | Threedimensional model preparation method, apparatus and system |
Non-Patent Citations (3)
| Title |
|---|
| 李燕等: "激光测距与卫星导航定位技术组合研究", 《激光杂志》 * |
| 王春光: "从数字化测绘到信息化测绘的测绘学科新进展", 《科技视界》 * |
| 肖玉钢等: "北斗卫星导航系统的毫米级精度变形监测算法与实现", 《测绘学报》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111768487A (en) * | 2020-06-11 | 2020-10-13 | 武汉市工程科学技术研究院 | Geological rock data three-dimensional reconstruction system and method based on three-dimensional point cloud base |
| CN111768487B (en) * | 2020-06-11 | 2023-11-28 | 武汉市工程科学技术研究院 | Geological formation data three-dimensional reconstruction system and method based on three-dimensional point cloud library |
| CN112489064A (en) * | 2020-12-14 | 2021-03-12 | 桂林电子科技大学 | Panorama segmentation method based on edge scaling correction |
| CN112489064B (en) * | 2020-12-14 | 2022-03-25 | 桂林电子科技大学 | Panorama segmentation method based on edge scaling correction |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109544677B (en) | Indoor scene main structure reconstruction method and system based on depth image key frame | |
| CN108509848B (en) | The real-time detection method and system of three-dimension object | |
| CN106803267B (en) | Kinect-based indoor scene three-dimensional reconstruction method | |
| CN107133325B (en) | Internet photo geographic space positioning method based on street view map | |
| Hu et al. | Approaches to large-scale urban modeling | |
| US10043097B2 (en) | Image abstraction system | |
| CN103839277B (en) | A kind of mobile augmented reality register method of outdoor largescale natural scene | |
| CN110135455A (en) | Image matching method, device and computer readable storage medium | |
| CN108052942B (en) | Visual image recognition method for aircraft flight attitude | |
| Wei et al. | Applications of structure from motion: a survey | |
| Pan et al. | Rapid scene reconstruction on mobile phones from panoramic images | |
| CN102959946A (en) | Augmenting image data based on related 3d point cloud data | |
| CN107767456A (en) | A kind of object dimensional method for reconstructing based on RGB D cameras | |
| Kim et al. | Interactive 3D building modeling method using panoramic image sequences and digital map | |
| Özbay et al. | A voxelize structured refinement method for registration of point clouds from Kinect sensors | |
| CN108010122B (en) | Method and system for reconstructing and measuring three-dimensional model of human body | |
| CN109118576A (en) | Large scene three-dimensional reconstruction system and method for reconstructing based on BDS location-based service | |
| Yuan et al. | 3D point cloud recognition of substation equipment based on plane detection | |
| Yin et al. | Virtual reconstruction method of regional 3D image based on visual transmission effect | |
| CN113902802A (en) | Visual positioning method and related device, electronic equipment and storage medium | |
| CN113362467A (en) | Point cloud preprocessing and ShuffleNet-based mobile terminal three-dimensional pose estimation method | |
| CN117132737A (en) | Three-dimensional building model construction method, system and equipment | |
| CN116894876A (en) | 6-DOF positioning method based on real-time image | |
| CN116843867A (en) | Augmented reality virtual-real fusion method, electronic device and storage medium | |
| Lee et al. | Determination of building model key points using multidirectional shaded relief images generated from airborne LiDAR data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190101 |