CN107767464A - A kind of augmented reality method for minimally invasive surgery - Google Patents
A kind of augmented reality method for minimally invasive surgery Download PDFInfo
- Publication number
- CN107767464A CN107767464A CN201710927040.4A CN201710927040A CN107767464A CN 107767464 A CN107767464 A CN 107767464A CN 201710927040 A CN201710927040 A CN 201710927040A CN 107767464 A CN107767464 A CN 107767464A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- frame
- key frame
- msubsup
- 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.)
- Withdrawn
Links
Landscapes
- Image Processing (AREA)
Abstract
A kind of augmented reality method for minimally invasive surgery proposed in the present invention, its main contents include:Mainly include landmark points detection, frame by frame pose estimation, the bundle adjustment based on key frame, intensive Stereo matching, enlarging geometric grid, its process is, the pose estimation of the video camera of each frame is carried out with matching system first by synchronous positioning, then 3 D stereo matching is carried out with rebuilding and being transformed into global surface mesh for each key frame, re-projection goes back to two-dimentional interface to last global surface mesh again, completes pixel conversion.The present invention can handle the human organ structure chart of different light conditions, there is provided an interactive geometrical layers solve the estimation of camera position, while improve the recognition effect and positioning accuracy of image.
Description
Technical field
The present invention relates to augmented reality field, more particularly, to a kind of augmented reality side for minimally invasive surgery
Method.
Background technology
Augmented reality is one and combines the multi-field knowledge such as computer technology, the communication technology, computer graphics
Technology, the purpose is to which additional information caused by computer is fused in the real world scene that user is seen, enhancing
Perception of the user to real scene.The advantages of being combined due to its actual situation, medical science, robot learning, military affairs, education,
There is a wide prospect in the fields such as manufacturing industry, for example, dangerous working environment is produced to human body, based on training place real estate
The virtual military target of life and strategic plan information, the manufacture of complicated machinery and maintenance etc. suffer from bigger potentiality to be exploited.
In medical domain, the human lenses 3-D graphic obtained by CT or MRI scan can be superimposed upon corresponding body part, had
Surgical procedure is instructed beneficial to doctor.Moreover, augmented reality can be additionally used in virtual human body internal anatomy, virtual human body work(
Energy, virtual surgery simulation, remote operation etc., especially, there is important application in terms of minimally invasive surgery.
Augmented reality be used for medical image integration technology there is a problem in that:1st, because image is in the process of collection
In influenceed by body surface reflection and ambient noise etc., the physical feature unobvious after image objects, be not easy to extract
And identification;2nd, because the influence for the factor that distorts, camera calibration have larger error in augmented reality image co-registration alignment system,
Therefore the three-dimensional coordinate precision of the spatial point obtained need further to improve;3rd, because operative site is different, in order to more preferable
Doctor, the position of patient are adapted to, the intellectuality of system platform and adaptability have much room for improvement.
The present invention proposes a kind of new frame that Stereo matching is carried out based on zero-mean regularization correlation Condition.Using same
Step positioning carries out the pose estimation of the video camera of each frame with matching system, then carries out 3 D stereo for each key frame
With rebuilding and being transformed into global surface mesh, re-projection goes back to two-dimentional interface to last global surface mesh again, completes pixel and turns for matching
Change.The present invention can handle the human organ structure chart of different light conditions, there is provided interactive geometrical layers solve to photograph
The estimation that seat in the plane is put, while improve the recognition effect and positioning accuracy of image.
The content of the invention
For solving the problems, such as to apply augmented reality in minimally invasive surgery, it is an object of the invention to provide a kind of pin
To the augmented reality method of minimally invasive surgery, it is proposed that a kind of to carry out the new of Stereo matching based on zero-mean regularization correlation Condition
Framework.
To solve the above problems, the present invention provides a kind of augmented reality method for minimally invasive surgery, its main contents bag
Include:
(1) landmark points detect;
(2) pose estimation frame by frame;
(3) bundle adjustment based on key frame;
(4) intensive Stereo matching;
(5) geometric grid is extended.
Wherein, described landmark points detection, corresponding left and right image is looked for using detector to feed back tracking video camera
Position, the coordinate on their x-axis direction are usedRepresent, then in the case of known focal length f and benchmark B, then image is each
Point can obtain to endoscope vertical range Z according to similar triangulation:
Wherein,Represent the distance of key point between left and right two images.
Described pose estimation frame by frame, including lap position renewal and swivel estimation.
Described lap position renewal, needs to track lap position frame by frame in real time, based on current in the application of enhancing technology
Linear velocity vtWith angular speed wt, after initialization after one section of of short duration time Δ t, use constant motion model estimated location rt+1
And quaternary number rotation qt+1:
Wherein, a and α is renewal coefficient respectively.
Described swivel estimation, using random sample aggregating algorithm, in each iteration, randomly selects current point setArrive
Next point setIn corresponding to three pairs of 3d viewpoints be used to calculate swivel matrix R and transposition T, its result is following by minimizing
Object function obtains:
Being found by each iteration can formula (3) intermediate value minimum R and T.
The described bundle adjustment based on key frame, including key frame screening and bundle adjustment calculate.
Described key frame screening, retains the frame with material impact to increase backdating capability and raising based on certain criterion
Robustness, while reduce the amount of calculation that global optimization is brought, the criterion is, after initialization, meets between any two frame (i.e.
Between key frame to be determined and a upper fixed key frame) common keypoints less than 80% and more than 50 condition person,
Key frame can be classified as.
Described bundle adjustment calculates, once it is determined that good key frame, then go renolation each using bundle adjustment
Key frame KFiWith landmark points PjThree-dimensional position, specifically, for the re-projection error between two-dimentional key point, pass through minimum
Total Hans Huber robustness loss function goes to obtain the three-dimensional position on camera angle, i.e.,:
Wherein, ρhIt is the constant factor of loss function, CamProj () represents video camera visual projection function.
Described intensive Stereo matching, landmark points are carried out with intensive reconstruct so that the change of environment is better described, according to left and right
Corresponding pixel p and its inconsistent distance d designs a loss function C in image, specifically,
Wherein,Refer to NpCriticize the average density put in image centered on pixel p.
Described enlarging geometric grid, uses transition matrix Tf2wThe estimate of three-dimensional point set is transformed into now from frame space
The real space, specifically, the surface mesh of point set is reconstructed first by triangulation method, then using the video camera by estimation
Posture goes to extend global surface mesh, and last image is projected back the two dimensional surface consistent with camera angles.
Brief description of the drawings
Fig. 1 is a kind of system flow chart of augmented reality method for minimally invasive surgery of the invention.
Fig. 2 is a kind of example schematic diagram of the measurement application of augmented reality method for minimally invasive surgery of the invention.
Fig. 3 is a kind of example schematic diagram of the other application of augmented reality method for minimally invasive surgery of the invention.
Embodiment
It should be noted that in the case where not conflicting, the feature in embodiment and embodiment in the application can phase
Mutually combine, the present invention is described in further detail with specific embodiment below in conjunction with the accompanying drawings.
Fig. 1 is a kind of system flow chart of augmented reality method for minimally invasive surgery of the invention.Mainly include landmark points
Detection;Pose estimation frame by frame;Bundle adjustment based on key frame;Intensive Stereo matching;Extend geometric grid.
Wherein, landmark points are detected, and corresponding left and right image is looked for feed back the position of tracking video camera using detector,
Coordinate on their x-axis direction is usedRepresent, then in the case of known focal length f and benchmark B, then image each point with it is interior
Sight glass vertical range Z can obtain according to similar triangulation:
Wherein,Represent the distance of key point between left and right two images.
Pose estimation frame by frame, including lap position renewal and swivel estimation.
Lap position updates, and needs to track lap position frame by frame in real time in the application of enhancing technology, based on current linear velocity
vtWith angular speed wt, after initialization after one section of of short duration time Δ t, use constant motion model estimated location rt+1And four
First number rotates qt+1:
Wherein, a and α is renewal coefficient respectively.
Swivel estimation, using random sample aggregating algorithm, in each iteration, randomly selects current point setTo subsequent point
CollectionIn corresponding to three pairs of 3d viewpoints be used to calculate swivel matrix R and transposition T, its result is by minimizing following target letter
Number obtains:
Being found by each iteration can formula (3) intermediate value minimum R and T.
Bundle adjustment based on key frame, including key frame screening and bundle adjustment calculate.
Key frame screens, and retains the frame with material impact based on certain criterion to increase backdating capability and improve robust
Property, while the amount of calculation that global optimization is brought is reduced, the criterion is, after initialization, meets (to treat really between any two frame
Between fixed key frame and a upper fixed key frame) common keypoints less than 80% and more than 50 condition person, can arrange
For key frame.
Bundle adjustment calculates, once it is determined that good key frame, then remove each key frame of renolation using bundle adjustment
KFiWith landmark points PjThree-dimensional position, specifically, for the re-projection error between two-dimentional key point, by minimizing total Hu shellfish
You go to obtain the three-dimensional position on camera angle robustness loss function, i.e.,:
Wherein, ρhIt is the constant factor of loss function, CamProj () represents video camera visual projection function.
Intensive Stereo matching, landmark points are carried out with intensive reconstruct so that the change of environment is better described, according in left images
Corresponding pixel p and its inconsistent distance d designs a loss function C, specifically,
Wherein,Refer to NpCriticize the average density put in image centered on pixel p.
Geometric grid is extended, uses transition matrix Tf2wThe estimate of three-dimensional point set is transformed into realistic space from frame space,
Specifically, the surface mesh of point set is reconstructed first by triangulation method, is then gone using the video camera posture by estimation
Global surface mesh is extended, last image is projected back the two dimensional surface consistent with camera angles.
Fig. 2 is a kind of example schematic diagram of the measurement application of augmented reality method for minimally invasive surgery of the invention.Such as figure
It is shown, it is observed that the measurement effect in two dimensional surface (figure of left column a, c two) is not directly perceived, by the enhancing skill of the present invention
After art application, the measurement of its tripleplane is more directly perceived and should be readily appreciated that.
Fig. 3 is a kind of example schematic diagram of the other application of augmented reality method for minimally invasive surgery of the invention.Such as figure
It is shown, it is observed that the figure of the first row a, b two illustrate how on human organ labelling and conversion diagram angle should
With;Second row c, d two figure illustrate how the human organ of the complex background of close light is subjected to contrast enhancing and highlighted
Change is handled, so as to reach the effect at positioning most critical position.
For those skilled in the art, the present invention is not restricted to the details of above-described embodiment, in the essence without departing substantially from the present invention
In the case of refreshing and scope, the present invention can be realized with other concrete forms.In addition, those skilled in the art can be to this hair
Bright to carry out various changes and modification without departing from the spirit and scope of the present invention, these improvement and modification also should be regarded as the present invention's
Protection domain.Therefore, appended claims are intended to be construed to include preferred embodiment and fall into all changes of the scope of the invention
More and change.
Claims (10)
- A kind of 1. augmented reality method for minimally invasive surgery, it is characterised in that mainly include landmark points detection (one);Appearance frame by frame Gesture estimates (two);Bundle adjustment (three) based on key frame;Intensive Stereo matching (four);Extend geometric grid (five).
- 2. detect (one) based on the landmark points described in claims 1, it is characterised in that using detector look for it is corresponding left, Right image to feed back the position of tracking video camera, use by the coordinate on their x-axis directionRepresent, then in known focal length f In the case of benchmark B, then image each point can obtain to endoscope vertical range Z according to similar triangulation:<mrow> <mfrac> <mrow> <mi>B</mi> <mo>-</mo> <mrow> <mo>(</mo> <msubsup> <mi>x</mi> <mi>P</mi> <mi>l</mi> </msubsup> <mo>-</mo> <msubsup> <mi>x</mi> <mi>P</mi> <mi>r</mi> </msubsup> <mo>)</mo> </mrow> </mrow> <mrow> <mi>Z</mi> <mo>-</mo> <mi>f</mi> </mrow> </mfrac> <mo>=</mo> <mfrac> <mi>B</mi> <mi>Z</mi> </mfrac> <mo>&DoubleRightArrow;</mo> <mi>Z</mi> <mo>=</mo> <mfrac> <mrow> <mi>f</mi> <mo>&CenterDot;</mo> <mi>B</mi> </mrow> <msub> <mi>d</mi> <mi>P</mi> </msub> </mfrac> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </mrow>Wherein,Represent the distance of key point between left and right two images.
- 3. based on the pose estimation frame by frame (two) described in claims 1, it is characterised in that update and turn including lap position Estimation.
- 4. based on described in claims 3 lap position renewal, it is characterised in that enhancing technology application in need in real time by Frame tracks lap position, based on current linear velocity vtWith angular speed wt, after initialization after one section of of short duration time Δ t, use Constant motion model estimated location rt+1And quaternary number rotation qt+1:<mrow> <mfenced open = "{" close = ""> <mtable> <mtr> <mtd> <mrow> <msub> <mi>r</mi> <mrow> <mi>t</mi> <mo>+</mo> <mn>1</mn> </mrow> </msub> <mo>=</mo> <msub> <mi>r</mi> <mi>t</mi> </msub> <mo>+</mo> <msub> <mi>v</mi> <mi>t</mi> </msub> <mo>&CenterDot;</mo> <mi>&Delta;</mi> <mi>t</mi> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msub> <mi>q</mi> <mrow> <mi>t</mi> <mo>+</mo> <mn>1</mn> </mrow> </msub> <mo>=</mo> <msub> <mi>q</mi> <mi>t</mi> </msub> <mo>&times;</mo> <mi>q</mi> <mrow> <mo>(</mo> <msub> <mi>w</mi> <mi>t</mi> </msub> <mo>&CenterDot;</mo> <mi>&Delta;</mi> <mi>t</mi> <mo>)</mo> </mrow> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msub> <mi>v</mi> <mi>t</mi> </msub> <mo>=</mo> <msub> <mi>v</mi> <mrow> <mi>t</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> <mo>+</mo> <msub> <mi>a</mi> <mrow> <mi>t</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> <mo>&CenterDot;</mo> <mi>&Delta;</mi> <mi>t</mi> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msub> <mi>w</mi> <mi>t</mi> </msub> <mo>=</mo> <msub> <mi>w</mi> <mrow> <mi>t</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> <mo>+</mo> <msub> <mi>&alpha;</mi> <mrow> <mi>t</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> <mo>&CenterDot;</mo> <mi>&Delta;</mi> <mi>t</mi> </mrow> </mtd> </mtr> </mtable> </mfenced> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </mrow>Wherein, a and α is renewal coefficient respectively.
- 5. based on the swivel estimation described in claims 3, it is characterised in that using random sample aggregating algorithm, changing every time Dai Zhong, randomly select current point setTo next point setIn corresponding to three pairs of 3d viewpoints be used to calculating swivel matrix R and Transposition T, its result are obtained by minimizing following object function:<mrow> <mi>min</mi> <munderover> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <mo>|</mo> <mo>|</mo> <msubsup> <mi>p</mi> <mi>t</mi> <mi>i</mi> </msubsup> <mo>-</mo> <mrow> <mo>(</mo> <mi>R</mi> <mo>*</mo> <msubsup> <mi>p</mi> <mrow> <mi>t</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>i</mi> </msubsup> <mo>+</mo> <mi>T</mi> <mo>)</mo> </mrow> <mo>|</mo> <mo>|</mo> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>3</mn> <mo>)</mo> </mrow> </mrow>Being found by each iteration can formula (3) intermediate value minimum R and T.
- 6. based on the bundle adjustment (three) based on key frame described in claims 1, it is characterised in that sieved including key frame Choosing and bundle adjustment calculate.
- 7. based on the key frame screening described in claims 6, it is characterised in that being retained based on certain criterion has material impact Frame to increase backdating capability and improve robustness, while reduce the amount of calculation that global optimization is brought, the criterion is, initial After change, meet that the common keypoints between any two frame (between key frame i.e. to be determined and a upper fixed key frame) are few In 80% and more than 50 condition person, key frame can be classified as.
- 8. calculated based on the bundle adjustment described in claims 6, it is characterised in that once it is determined that good key frame, then utilize Bundle adjustment removes each key frame KF of renolationiWith landmark points PjThree-dimensional position, specifically, for two-dimentional key point it Between re-projection error, go to obtain the three-dimensional position on camera angle by minimizing total Hans Huber robustness loss function Put, i.e.,:<mrow> <mi>arg</mi> <munder> <mrow> <mi>m</mi> <mi>i</mi> <mi>n</mi> </mrow> <mrow> <msub> <mi>KF</mi> <mi>i</mi> </msub> <mo>,</mo> <msub> <mi>P</mi> <mi>j</mi> </msub> </mrow> </munder> <munder> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </munder> <msub> <mi>&rho;</mi> <mi>h</mi> </msub> <mrow> <mo>(</mo> <mo>|</mo> <mo>|</mo> <msub> <mi>p</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>-</mo> <mi>C</mi> <mi>a</mi> <mi>m</mi> <mi>P</mi> <mi>r</mi> <mi>o</mi> <mi>j</mi> <mo>(</mo> <mrow> <msub> <mi>KF</mi> <mi>i</mi> </msub> <mo>,</mo> <msub> <mi>P</mi> <mi>j</mi> </msub> </mrow> <mo>)</mo> <mo>|</mo> <mo>|</mo> <mo>)</mo> </mrow> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>4</mn> <mo>)</mo> </mrow> </mrow>Wherein, ρhIt is the constant factor of loss function, CamProj () represents video camera visual projection function.
- 9. based on the intensive Stereo matching (four) described in claims 1, it is characterised in that to landmark points carry out it is intensive reconstruct with The change of environment is better described, a loss function is designed according to corresponding pixel p in left images and its inconsistent distance d C, specifically,<mrow> <mi>C</mi> <mrow> <mo>(</mo> <mi>p</mi> <mo>,</mo> <mi>d</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msub> <mo>&Sigma;</mo> <mrow> <mi>q</mi> <mo>&Element;</mo> <msub> <mi>N</mi> <mi>p</mi> </msub> </mrow> </msub> <mrow> <mo>(</mo> <msub> <mi>I</mi> <mi>L</mi> </msub> <mo>(</mo> <mi>q</mi> <mo>)</mo> <mo>-</mo> <msub> <mover> <mi>I</mi> <mo>&OverBar;</mo> </mover> <mi>L</mi> </msub> <mo>(</mo> <mi>q</mi> <mo>)</mo> <mo>)</mo> </mrow> <mo>&CenterDot;</mo> <mrow> <mo>(</mo> <msub> <mi>I</mi> <mi>R</mi> </msub> <mo>(</mo> <mrow> <mi>q</mi> <mo>-</mo> <mi>d</mi> </mrow> <mo>)</mo> <mo>-</mo> <msub> <mover> <mi>I</mi> <mo>&OverBar;</mo> </mover> <mi>R</mi> </msub> <mo>(</mo> <mrow> <mi>q</mi> <mo>-</mo> <mi>d</mi> </mrow> <mo>)</mo> <mo>)</mo> </mrow> </mrow> <msqrt> <mrow> <msub> <mo>&Sigma;</mo> <mrow> <mi>q</mi> <mo>&Element;</mo> <msub> <mi>N</mi> <mi>p</mi> </msub> </mrow> </msub> <msup> <mrow> <mo>(</mo> <msub> <mi>I</mi> <mi>L</mi> </msub> <mo>(</mo> <mi>q</mi> <mo>)</mo> <mo>-</mo> <msub> <mover> <mi>I</mi> <mo>&OverBar;</mo> </mover> <mi>L</mi> </msub> <mo>(</mo> <mi>q</mi> <mo>)</mo> <mo>)</mo> </mrow> <mn>2</mn> </msup> <mo>&CenterDot;</mo> <msub> <mo>&Sigma;</mo> <mrow> <mi>q</mi> <mo>&Element;</mo> <msub> <mi>N</mi> <mi>p</mi> </msub> </mrow> </msub> <msup> <mrow> <mo>(</mo> <msub> <mi>I</mi> <mi>R</mi> </msub> <mo>(</mo> <mrow> <mi>q</mi> <mo>-</mo> <mi>d</mi> </mrow> <mo>)</mo> <mo>-</mo> <msub> <mover> <mi>I</mi> <mo>&OverBar;</mo> </mover> <mi>R</mi> </msub> <mo>(</mo> <mrow> <mi>q</mi> <mo>-</mo> <mi>d</mi> </mrow> <mo>)</mo> <mo>)</mo> </mrow> <mn>2</mn> </msup> </mrow> </msqrt> </mfrac> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>5</mn> <mo>)</mo> </mrow> </mrow>Wherein,Refer to NpCriticize the average density put in image centered on pixel p.
- 10. based on the enlarging geometric grid (five) described in claims 1, it is characterised in that use transition matrix Tf2wBy three-dimensional The estimate of point set is transformed into realistic space from frame space, specifically, the surface of point set is reconstructed first by triangulation method Grid, then go to extend global surface mesh using the video camera posture by estimation, last image is projected back and video camera The consistent two dimensional surface in visual angle.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710927040.4A CN107767464A (en) | 2017-10-09 | 2017-10-09 | A kind of augmented reality method for minimally invasive surgery |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710927040.4A CN107767464A (en) | 2017-10-09 | 2017-10-09 | A kind of augmented reality method for minimally invasive surgery |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107767464A true CN107767464A (en) | 2018-03-06 |
Family
ID=61266554
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710927040.4A Withdrawn CN107767464A (en) | 2017-10-09 | 2017-10-09 | A kind of augmented reality method for minimally invasive surgery |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107767464A (en) |
-
2017
- 2017-10-09 CN CN201710927040.4A patent/CN107767464A/en not_active Withdrawn
Non-Patent Citations (1)
Title |
---|
LONG CHEN等: "Real-time Geometry-Aware Augmented Reality in Minimally Invasive Surgery", 《ARXIV:1708.01234V1》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5877053B2 (en) | Posture estimation apparatus and posture estimation method | |
ES2812578T3 (en) | Estimating a posture based on silhouette | |
WO2020054442A1 (en) | Articulation position acquisition method and device, and motion acquisition method and device | |
EP2739036B1 (en) | Plane-characteristic-based markerless augmented reality system and method for operating same | |
CN102609942B (en) | Depth map is used to carry out mobile camera location | |
CN104995666B (en) | Method for indicating virtual information in true environment | |
US20170140552A1 (en) | Apparatus and method for estimating hand position utilizing head mounted color depth camera, and bare hand interaction system using same | |
Canessa et al. | Calibrated depth and color cameras for accurate 3D interaction in a stereoscopic augmented reality environment | |
CN103543830B (en) | Method for mapping human skeleton points to virtual three-dimensional space points in three-dimensional display | |
CN102510506B (en) | Virtual and real occlusion handling method based on binocular image and range information | |
WO2018019272A1 (en) | Method and apparatus for realizing augmented reality on the basis of plane detection | |
JP7423683B2 (en) | image display system | |
JP2011238222A5 (en) | ||
JP7427188B2 (en) | 3D pose acquisition method and device | |
US11436790B2 (en) | Passthrough visualization | |
Rodas et al. | See it with your own eyes: Markerless mobile augmented reality for radiation awareness in the hybrid room | |
EP3185212B1 (en) | Dynamic particle filter parameterization | |
CN110555869A (en) | method and system for extracting primary and secondary motion in augmented reality systems | |
CN106683163A (en) | Imaging method and system used in video monitoring | |
CN111489392B (en) | Single target human motion posture capturing method and system in multi-person environment | |
Chen et al. | Camera networks for healthcare, teleimmersion, and surveillance | |
Zou et al. | Automatic reconstruction of 3D human motion pose from uncalibrated monocular video sequences based on markerless human motion tracking | |
CN104732586B (en) | A kind of dynamic body of 3 D human body and three-dimensional motion light stream fast reconstructing method | |
CN103260008B (en) | A kind of image position is to the projection conversion method of physical location | |
CN102663812A (en) | Direct method of three-dimensional motion detection and dense structure reconstruction based on variable optical flow |
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 | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20180306 |
|
WW01 | Invention patent application withdrawn after publication |