CN110148216A - A kind of method of double ball curtain cameras and double ball curtain camera three-dimensional modelings - Google Patents
A kind of method of double ball curtain cameras and double ball curtain camera three-dimensional modelings Download PDFInfo
- Publication number
- CN110148216A CN110148216A CN201910437479.8A CN201910437479A CN110148216A CN 110148216 A CN110148216 A CN 110148216A CN 201910437479 A CN201910437479 A CN 201910437479A CN 110148216 A CN110148216 A CN 110148216A
- Authority
- CN
- China
- Prior art keywords
- ball curtain
- camera
- photo
- auxiliary
- main
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/55—Depth or shape recovery from multiple images
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P10/00—Technologies related to metal processing
- Y02P10/25—Process efficiency
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Studio Devices (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a kind of double ball curtain camera three-dimensional modeling methods and double ball curtain cameras, which comprises S1, data acquisition are shot using double ball curtain cameras, obtain the main photo and auxiliary photo of same shooting time;S2 is extracted characteristic point data to main photo and auxiliary photo respectively using SIFT algorithm, matched to the characteristic point of main photo and auxiliary photo, and calculated the depth data of main photo eigen point using binocular vision algorithm;S3, characteristic point data and depth data based on main photo calculate camera position and dense point cloud using binocular SFM algorithm;S4 carries out structured modeling to camera position and dense point cloud, obtains the ball curtain photo of one group of main phase machine;S5, the ball curtain photo based on each group of main phase machine carry out textures to model.The present invention, instead of depth camera, is reduced costs by way of double ball curtains, improves modeling speed and effect.
Description
Technical field
The present invention relates to three-dimension scene simulation technical fields, and in particular to a kind of double ball curtain camera three-dimensional modeling methods and double
Ball curtain camera.
Background technique
In the prior art using ball curtain camera obtain be ball curtain photo two-dimensional signal, the depth of ball curtain photo can not be obtained
Information is spent, such Modeling Calculation is complicated, and modeling speed is slow, effect is also poor.
Summary of the invention
Based on the deficiencies of the prior art, the present invention is while providing a kind of double ball curtain camera three-dimensional modeling methods, also
A kind of double ball curtain cameras are provided, lacking depth value solving traditional ball curtain photo containing only two-dimensional signal leads to modeling speed
Slowly, poor, the at high cost problem of effect.
The technical solution of the present invention is as follows:
A kind of double ball curtain camera three-dimensional modeling methods, which is characterized in that the described method includes:
S1, data acquisition, is shot using double ball curtain cameras, obtains the main photo and auxiliary photo of same shooting time;
S2 extracts characteristic point data to main photo and auxiliary photo respectively using SIFT algorithm, shines main photo and auxiliary
The characteristic point of piece is matched, and the depth data of main photo eigen point is calculated using binocular vision algorithm;
S3, characteristic point data and depth data based on main photo calculate camera position and dense using binocular SFM algorithm
Point cloud;
S4 carries out structured modeling to camera position and dense point cloud, obtains the ball curtain photo of one group of main phase machine;
S5, the ball curtain photo based on each group of main phase machine carry out textures to model.
Further, in S1 to major-minor photo extract characteristic point data extract be characteristic point 2-D data.
Further, three dimensions of the characteristic point and the depth characteristic point data and depth data of main photo extracted in S2
According to.
Further, in S1 the cue ball curtain photo of same shooting time and auxiliary ball curtain photo be based on double ball curtain cameras with
The photo that the peace distance of auxiliary ball curtain camera is shot in the range of being 10-20cm.
Further, in S1 the cue ball curtain photo of same shooting time and auxiliary ball curtain photo be based on double ball curtain cameras with
The photo that the distance between auxiliary ball curtain camera is shot in the range of being 15cm.
A kind of double ball curtain cameras, it is characterised in that: including lens bracket, one end of the lens bracket is provided with cue ball curtain
Camera, the other end of the lens bracket are provided with auxiliary ball curtain camera, and the cue ball curtain camera is connected with the auxiliary ball curtain camera
Same image processor, described image processor are used for the image of the cue ball curtain camera and the auxiliary ball curtain camera
Reason and splicing, wherein the upper end of the lens bracket is arranged in cue ball curtain camera, and auxiliary ball curtain camera is arranged in the lens bracket
Lower end.
Further, the cue ball curtain camera includes at least two main lens, and the auxiliary ball curtain camera includes at least two
Auxiliary camera lens, plane locating for the light mandrel of all main lens are parallel with plane locating for the light mandrel of all auxiliary camera lenses.
Further, the cue ball curtain camera includes two main lens, and two main lens are backwards to setting in the mirror
The upper end of head bracket;The auxiliary ball curtain camera includes two attachment lenses, and two auxiliary camera lenses are backwards to setting in the lens bracket
Lower end, the light mandrel of two main lens is overlapped, and the light mandrel of two auxiliary camera lenses is overlapped, the light mandrel of two main lens and two
The light mandrel of auxiliary camera lens is vertical.
Further, the cue ball curtain camera includes four main lens, four main lens backwards to and right-angled intersection
The upper end of the lens bracket is set;The auxiliary ball curtain camera includes four attachment lenses, four auxiliary camera lenses backwards to and cross
Lower end arranged in a crossed manner in the lens bracket, projection of four main lens in main lens mounting surface and four auxiliary camera lenses are in master
Projection in camera lens mounting surface is overlapped.
Further, the distance range of cue ball curtain camera and auxiliary ball curtain camera is 10~20cm.
The invention has the benefit that this method and double ball curtain cameras can be in the cue ball curtain photographs of same level
The shooting of piece shooting and auxiliary ball curtain photo makes each shooting of same time obtain two ball curtain photos, i.e., cue ball curtain photo with
Auxiliary ball curtain photo, two photos have the visual difference of 10-20cm, and the distance of visual difference is true according to the spacing of major-minor ball curtain camera
It is fixed, instead of depth camera by way of double ball curtains, reduce costs, meanwhile, by the way of three-dimensional modeling, assisting
With the help of photo, the depth value of main photo eigen point is directly calculated, so that each group of main photo eigen point obtains three dimensions
According to this makes matching result more accurate, and modeling effect is more stable.In addition, comparing conventional method and dress in data acquisition
It sets, reduces data collection capacity still and can achieve stable modelling effect, reduce calculation amount, when reducing data processing
Between.
Detailed description of the invention
Fig. 1 is flow chart of the method for the present invention;
Fig. 2 is the structural schematic diagram of double ball curtain cameras of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
As shown in Figure 1, a kind of double ball curtain camera three-dimensional modeling methods, which comprises
S1, data acquisition, is shot using double ball curtain cameras, obtains the main photo and auxiliary photo of same shooting time;
S2 extracts characteristic point data to main photo and auxiliary photo respectively using SIFT algorithm, shines main photo and auxiliary
The characteristic point of piece is matched, and the depth data of main photo eigen point is calculated using binocular vision algorithm;Wherein, to major-minor
What photo extracted characteristic point data extraction is the 2-D data of characteristic point.
S3, characteristic point data and depth data based on main photo calculate camera position and thicker using binocular SFM algorithm
Close point cloud;Wherein, the characteristic point data to main photo and depth data extract the three-dimensional data of the characteristic point and depth that are.
S4 carries out structured modeling to camera position and denser point cloud, obtains the ball curtain photo of one group of main phase machine;
S5, the ball curtain photo based on each group of main phase machine carry out textures to model.
Wherein, after the depth data that main photo is obtained in S3, so that it may by 2D characteristic point before, be converted into 3D feature
Point, in this way, when carrying out camera calibration, so that it may will constrain from single ball purpose re-projection two-dimensional pixel error of coordinate (2D)
Become the three dimensional space coordinate error (3D) of double ball purpose re-projections.In this way, carrying out a dimension than traditional having more when calculating
Information can make calculating process quicker, and calculated result is more stable.
The cue ball curtain photo of same shooting time and auxiliary ball curtain photo are based on cue ball curtain phase in double ball curtain cameras in S1
The photo that the distance between machine and auxiliary ball curtain camera are shot in the range of being 10-20cm, more preferably distance is the range of 15cm
The photo of interior shooting.
10~the 20cm that is staggered does the effect that spatial digitalized modeling is equivalent to binocular vision.Binocular vision when taking 15cm
Effect is best.
This method is directed to not in the shooting of the cue ball curtain photograph taking of same level and auxiliary ball curtain photo, makes the same time
It is each shooting obtain two ball curtain photos, i.e. cue ball curtain photo and auxiliary ball curtain photo, two photos has the vision of 10-20cm
The distance of difference, visual difference is determined according to the spacing of major-minor ball curtain camera, instead of depth camera by way of double ball curtains, is reduced
Cost, meanwhile, by the way of three-dimensional modeling, with the help of assisting photo, directly calculate main photo eigen point
Depth value, so that each group of main photo eigen point obtains three-dimensional data, this makes matching result more accurate, and modeling effect is more
Stablize.In addition, comparing conventional method in data acquisition, reducing data collection capacity still can achieve stable modelling effect,
Reduce calculation amount, reduces data processing time.
A kind of double ball curtain cameras, including lens bracket 1, one end of the lens bracket 1 are provided with cue ball curtain camera 2, institute
The other end for stating lens bracket 1 is provided with auxiliary ball curtain camera 3, and the cue ball curtain camera 2 and the auxiliary ball curtain camera 3 connect same
Image processor, described image processor be used for the image of the cue ball curtain camera and the auxiliary ball curtain camera carry out processing and
Splicing, wherein the upper end of the lens bracket 1 is arranged in cue ball curtain camera 2, and auxiliary ball curtain camera 3 is arranged in the lens bracket 1
Lower end.
And the embodiment of the present invention lens bracket 1 uses rectangular core structure.
As shown in Fig. 2, two camera lenses, i.e. the first main lens 21 and second all can be used in cue ball curtain camera and auxiliary ball curtain camera
The rectangular core structure is arranged in main lens 22, i.e., the first auxiliary camera lens 31 and the second auxiliary camera lens 32, two main lens 2
The rectangular core structure is arranged in the outside in two opposite faces of the upper end of lens bracket, two attachment lenses 3
The outside in two opposite faces of the lower end of lens bracket.The light mandrel of two main lens is overlapped, the light mandrel of two auxiliary camera lenses
It is overlapped, the light mandrel of two main lens is vertical with the light mandrel of two auxiliary camera lenses.The light mandrel and two auxiliary mirrors of two main lens
Head light mandrel it is vertical, two-by-two camera lens be in 90 degree, for ball curtain photo, the light core pair of every group of camera lens very accurately, have
Conducive to image mosaic, restore ball curtain photo.
The range of vertical range between the light mandrel of two main lens and the light mandrel of two auxiliary camera lenses is 10~20cm.
Vertical range between the light mandrel of two main lens of the present embodiment and the light mandrel of two auxiliary camera lenses is 15cm, it is staggered 10~
20cm does the effect that spatial digitalized modeling is equivalent to binocular vision.Binocular vision effect when taking 15cm is best, meanwhile, it is wrong
90 degree are opened, allows this four camera lenses that can carry out photo splicing, forms ball curtain photo.
Four main lens also can be used in the cue ball curtain camera, and four main lens are backwards and the setting of right-angled intersection exists
The upper end of the lens bracket of rectangular cartridge type, the auxiliary ball curtain camera also can be used four attachment lenses, four auxiliary camera lenses backwards to and
Right-angled intersection is arranged in the lower end of the lens bracket of rectangular cartridge type, projection of four main lens in main lens mounting surface with four
Projection of the auxiliary camera lens in main lens mounting surface is overlapped.Remaining feature is identical as the feature of two main lens and two auxiliary ends,
Each four camera lenses up and down, it is more preferable to the splicing effect of ball curtain photo.
It although an embodiment of the present invention has been shown and described, for the ordinary skill in the art, can be with
A variety of variations, modification, replacement can be carried out to these embodiments without departing from the principles and spirit of the present invention by understanding
And modification, the scope of the present invention is defined by the appended.
Claims (10)
1. a kind of double ball curtain camera three-dimensional modeling methods, which is characterized in that the described method includes:
S1, data acquisition, is shot using double ball curtain cameras, obtains the main photo and auxiliary photo of same shooting time;
S2 extracts characteristic point data to main photo and auxiliary photo respectively using SIFT algorithm, to main photo and assists photo
Characteristic point is matched, and the depth data of main photo eigen point is calculated using binocular vision algorithm;
S3, characteristic point data and depth data based on main photo calculate camera position and dense point using binocular SFM algorithm
Cloud;
S4 carries out structured modeling to camera position and dense point cloud, obtains the ball curtain photo of one group of main phase machine;
S5, the ball curtain photo based on each group of main phase machine carry out textures to model.
2. double ball curtain camera three-dimensional modeling methods as described in claim 1, it is characterised in that: extracted in S1 to major-minor photo special
What sign point data was extracted is the 2-D data of characteristic point.
3. double ball curtain camera three-dimensional modeling methods as claimed in claim 3, it is characterised in that: to the characteristic point of main photo in S2
The three-dimensional data of characteristic point and depth that data and depth data extract.
4. double ball curtain camera three-dimensional modeling methods as described in claim 1, it is characterised in that: the master of same shooting time in S1
Ball curtain photo and auxiliary ball curtain photo are to be based on the distance between cue ball curtain camera and auxiliary ball curtain camera in double ball curtain cameras
The photo shot in the range of 10-20cm.
5. double ball curtain camera three-dimensional modeling methods as claimed in claim 4, it is characterised in that: the master of same shooting time in S1
Ball curtain photo and auxiliary ball curtain photo are to be based on the distance between cue ball curtain camera and auxiliary ball curtain camera in double ball curtain cameras
The photo shot in the range of 15cm.
6. a kind of double ball curtain cameras, it is characterised in that: including lens bracket, one end of the lens bracket is provided with cue ball curtain phase
Machine, the other end of the lens bracket are provided with auxiliary ball curtain camera, and the cue ball curtain camera connects together with the auxiliary ball curtain camera
One image processor, described image processor is for handling the image of the cue ball curtain camera and the auxiliary ball curtain camera
And splicing, wherein the upper end of the lens bracket is arranged in cue ball curtain camera, and the lens bracket is arranged in auxiliary ball curtain camera
Lower end.
7. double ball curtain cameras as claimed in claim 6, it is characterised in that: the cue ball curtain camera includes at least two primary mirrors
Head, the auxiliary ball curtain camera include at least two auxiliary camera lenses, plane locating for the light mandrel of all main lens and all auxiliary camera lenses
Light mandrel locating for plane it is parallel.
8. ball curtain cameras as claimed in claim 7 double, it is characterised in that: the cue ball curtain camera includes two main lens, and two
The upper end of the lens bracket is arranged in a main lens backwards;The auxiliary ball curtain camera include two attachment lenses, two
The lower end of the lens bracket is arranged in auxiliary camera lens backwards, and the light mandrel of two main lens is overlapped, the light mandrel of two auxiliary camera lenses
It is overlapped, the light mandrel of two main lens is vertical with the light mandrel of two auxiliary camera lenses.
9. ball curtain cameras as claimed in claim 7 double, it is characterised in that: the cue ball curtain camera includes four main lens, and four
A main lens backwards to and right-angled intersection the upper end that the lens bracket is set;The auxiliary ball curtain camera includes four auxiliary
Help camera lens, four auxiliary camera lenses backwards to and right-angled intersection the lower end of the lens bracket is set, four main lens are pacified in main lens
Projection in dress face is overlapped with projection of four auxiliary camera lenses in main lens mounting surface.
10. double ball curtain cameras as described in claim 6-9, it is characterised in that: cue ball curtain camera with auxiliary ball curtain camera
Distance range is 10~20cm.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910437479.8A CN110148216B (en) | 2019-05-24 | 2019-05-24 | Three-dimensional modeling method of double-dome camera |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910437479.8A CN110148216B (en) | 2019-05-24 | 2019-05-24 | Three-dimensional modeling method of double-dome camera |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110148216A true CN110148216A (en) | 2019-08-20 |
CN110148216B CN110148216B (en) | 2023-03-24 |
Family
ID=67592989
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910437479.8A Active CN110148216B (en) | 2019-05-24 | 2019-05-24 | Three-dimensional modeling method of double-dome camera |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110148216B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112132466A (en) * | 2020-09-24 | 2020-12-25 | 中德(珠海)人工智能研究院有限公司 | Route planning method, device and equipment based on three-dimensional modeling and storage medium |
US11195297B2 (en) | 2019-08-29 | 2021-12-07 | China-Germany(Zhuhai)Artificial Intelligence Institute Co., Ltd | Method and system for visual localization based on dual dome cameras |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106908040A (en) * | 2017-03-06 | 2017-06-30 | 哈尔滨工程大学 | A kind of binocular panorama visual robot autonomous localization method based on SURF algorithm |
US20180035047A1 (en) * | 2016-07-29 | 2018-02-01 | Multimedia Image Solution Limited | Method for stitching together images taken through fisheye lens in order to produce 360-degree spherical panorama |
CN107666606A (en) * | 2016-07-29 | 2018-02-06 | 东南大学 | Binocular panoramic picture acquisition methods and device |
CN108288292A (en) * | 2017-12-26 | 2018-07-17 | 中国科学院深圳先进技术研究院 | A kind of three-dimensional rebuilding method, device and equipment |
CN108556739A (en) * | 2018-03-30 | 2018-09-21 | 东南大学 | Vehicle early warning device based on binocular full-view stereo vision |
CN108564617A (en) * | 2018-03-22 | 2018-09-21 | 深圳岚锋创视网络科技有限公司 | Three-dimensional rebuilding method, device, VR cameras and the panorama camera of more mesh cameras |
CN108629829A (en) * | 2018-03-23 | 2018-10-09 | 中德(珠海)人工智能研究院有限公司 | The three-dimensional modeling method and system that one bulb curtain camera is combined with depth camera |
CN109102537A (en) * | 2018-06-25 | 2018-12-28 | 中德人工智能研究院有限公司 | A kind of three-dimensional modeling method and system of laser radar and the combination of ball curtain camera |
CN109254579A (en) * | 2017-07-14 | 2019-01-22 | 上海汽车集团股份有限公司 | A kind of binocular vision camera hardware system, 3 D scene rebuilding system and method |
-
2019
- 2019-05-24 CN CN201910437479.8A patent/CN110148216B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180035047A1 (en) * | 2016-07-29 | 2018-02-01 | Multimedia Image Solution Limited | Method for stitching together images taken through fisheye lens in order to produce 360-degree spherical panorama |
CN107666606A (en) * | 2016-07-29 | 2018-02-06 | 东南大学 | Binocular panoramic picture acquisition methods and device |
CN106908040A (en) * | 2017-03-06 | 2017-06-30 | 哈尔滨工程大学 | A kind of binocular panorama visual robot autonomous localization method based on SURF algorithm |
CN109254579A (en) * | 2017-07-14 | 2019-01-22 | 上海汽车集团股份有限公司 | A kind of binocular vision camera hardware system, 3 D scene rebuilding system and method |
CN108288292A (en) * | 2017-12-26 | 2018-07-17 | 中国科学院深圳先进技术研究院 | A kind of three-dimensional rebuilding method, device and equipment |
CN108564617A (en) * | 2018-03-22 | 2018-09-21 | 深圳岚锋创视网络科技有限公司 | Three-dimensional rebuilding method, device, VR cameras and the panorama camera of more mesh cameras |
CN108629829A (en) * | 2018-03-23 | 2018-10-09 | 中德(珠海)人工智能研究院有限公司 | The three-dimensional modeling method and system that one bulb curtain camera is combined with depth camera |
CN108556739A (en) * | 2018-03-30 | 2018-09-21 | 东南大学 | Vehicle early warning device based on binocular full-view stereo vision |
CN109102537A (en) * | 2018-06-25 | 2018-12-28 | 中德人工智能研究院有限公司 | A kind of three-dimensional modeling method and system of laser radar and the combination of ball curtain camera |
Non-Patent Citations (1)
Title |
---|
崔岩: "球幕相机对大型建筑物及场景的三维数字化及其展示手段", 《东南文化》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11195297B2 (en) | 2019-08-29 | 2021-12-07 | China-Germany(Zhuhai)Artificial Intelligence Institute Co., Ltd | Method and system for visual localization based on dual dome cameras |
CN112132466A (en) * | 2020-09-24 | 2020-12-25 | 中德(珠海)人工智能研究院有限公司 | Route planning method, device and equipment based on three-dimensional modeling and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN110148216B (en) | 2023-03-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN100468465C (en) | Stereo vision three-dimensional human face modelling approach based on dummy image | |
CN108615244B (en) | A kind of image depth estimation method and system based on CNN and depth filter | |
CN100428805C (en) | Video camera reference method only using plane reference object image | |
CN109242954B (en) | Multi-view three-dimensional human body reconstruction method based on template deformation | |
CN109165680B (en) | Single-target object dictionary model improvement method in indoor scene based on visual SLAM | |
Furukawa et al. | Accurate, dense, and robust multiview stereopsis | |
CN107680159B (en) | Space non-cooperative target three-dimensional reconstruction method based on projection matrix | |
WO2024045632A1 (en) | Binocular vision and imu-based underwater scene three-dimensional reconstruction method, and device | |
CN107240129A (en) | Object and indoor small scene based on RGB D camera datas recover and modeling method | |
CN110310331B (en) | Pose estimation method based on combination of linear features and point cloud features | |
CN104036542B (en) | Spatial light clustering-based image surface feature point matching method | |
CN109859272A (en) | A kind of auto-focusing binocular camera scaling method and device | |
CN110956661B (en) | Method for calculating dynamic pose of visible light and infrared camera based on bidirectional homography matrix | |
CN109470149B (en) | Method and device for measuring position and posture of pipeline | |
CN107358633A (en) | Join scaling method inside and outside a kind of polyphaser based on 3 points of demarcation things | |
WO2018053952A1 (en) | Video image depth extraction method based on scene sample library | |
CN106017327A (en) | Structural light measurement sensor calibration method | |
CN110148216A (en) | A kind of method of double ball curtain cameras and double ball curtain camera three-dimensional modelings | |
WO2019201336A1 (en) | Light field based reflection removal | |
CN112150518B (en) | Attention mechanism-based image stereo matching method and binocular device | |
CN104182968A (en) | Method for segmenting fuzzy moving targets by wide-baseline multi-array optical detection system | |
CN110425983A (en) | A kind of monocular vision three-dimensional reconstruction distance measuring method based on polarization multi-spectrum | |
CN114022542A (en) | Three-dimensional reconstruction-based 3D database manufacturing method | |
CN111105451B (en) | Driving scene binocular depth estimation method for overcoming occlusion effect | |
CN111429571A (en) | Rapid stereo matching method based on spatio-temporal image information joint correlation |
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 |