image-based modelingvision.cse.psu.edu/research/symmcomp/workshop/ibm-longquan.pdfmodeling) objects...
TRANSCRIPT
![Page 1: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/1.jpg)
ImageImage--based Modelingbased Modeling
Long [email protected]
with contributions from T. Fang, M. Lhuillier, P. Tan, Y. Wei, J. Xiao, G. Zeng and P. Zhao
The department of Computer Science & Engineering, HKUST
![Page 2: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/2.jpg)
OverviewIntroductionFrom pixels to 3D points
Structure from motionA quasi-dense approach
From 3D points to objectsSmall-scale objects
Smooth surfaces, Hairs, TreesLarge-scale buildings
Façade, Buildings, Cities
Conclusions
![Page 3: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/3.jpg)
IntroductionComputer vision
3D reconstruction from imagesOne of the fundamental vision problems
Computer graphicsModeling: creating 3D digital models
Interactive methods (using softwares like Maya)Use sanners
Use photographs image-based modelingFor existing objectsPhotorealistic modelsCan be automated
![Page 4: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/4.jpg)
TV/movie post-production, special effects, Virtual realities, 3D game creationPhotogrammetryRobotics navigation…Modeling the world:
Google Earth, Virtual Earth!
Modeling/3D reconstruction applications
![Page 5: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/5.jpg)
One example of small-scale object
Rendered by Maya3D mesh modeltake images(1 of 35)
![Page 6: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/6.jpg)
One example of large-scale object
![Page 7: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/7.jpg)
How to get 3D points and camera positions from 2D pixels? (geometry computation or structure-from-motion (SFM))
How to get 3D objects from unstructured 3D points? (object modeling)
Objects are of different natures:Smooth objects (surface reconstruction)Curvilinear objects (hair modeling)Complex fine details (tree modeling)Large-scale objects (city modeling)
Two challenging components:
Smooth surface
Curves
Open surface patchTexturesRegular objects
![Page 8: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/8.jpg)
Structure from MotionStructure from Motion
![Page 9: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/9.jpg)
Standard Sparse SFMSparse feature detection (points of interest, or SIFT)Heuristic Initial matching of feature points by local image similaritiesCorrespondence using RANSAC with minimum algebraic solutionAuto-calibrationOptimization of cameras and points (bundle adjustment)
There are textbooks (Faugeras,Hartley/Zisserman) and commercial softwares for sparse SFM.
captured 51 images
Sparse 3D points and camera positions
……
![Page 10: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/10.jpg)
A Quasi-dense approach
Dense (per pixel) reconstruction is impossible Standard Sparse (feature) reconstruction is insufficient
Quasi-dense: re-sampled (un-interpolated) disparity maps
captured 26 images
3D points and camera positions
… …
A Quasi-Dense Approach to Surface Reconstruction from Uncalibrated ImagesLhuillier and Quan, IEEE PAMI 2005
![Page 11: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/11.jpg)
Advantages of Quasi-dense:
Hand-held camera (uncalibrated)Fully automatic computationRobust and accurate (the example is impossible with other existing methods)Pave the way for further object modeling (with sufficient density of points cloud, often impossible with sparse)
![Page 12: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/12.jpg)
Object ModelingObject Modeling
![Page 13: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/13.jpg)
Modeling
SFM
Structuring points into objects!
![Page 14: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/14.jpg)
Surface reconstruction
Model a wide variety of relatively smooth objectsPiror is ‘smoothness’Mathematically well studied:
Implicit surfaceVariational principle Level-set implementation
![Page 15: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/15.jpg)
Examples
![Page 16: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/16.jpg)
Hair Modeling
Difficult due to insufficient image resolution
Typically 500*500 pixels for 50,000 hair fibers
Only Hair strands, group of hair fibers, are visible
![Page 17: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/17.jpg)
Hair modeling approachPrior model: a curve approximated by a polylineConstraint: Polyline Growth by following edge orientation
![Page 18: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/18.jpg)
Modeling Hair from Images,Wei, Ofek, Quan, Shum, Siggraph 2004.
Examples, fiber by fiber, 300K fibers
![Page 19: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/19.jpg)
Tree modeling
Trees are ubiquitous but difficult to modelComplex geometry and topologyFine texture details
relative leaf size
Tree spectrum
![Page 20: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/20.jpg)
But, trees have self-similar structure: a natural generative model for both leaves and branchesAll leaves are similar: a generic model for all leaves of a type of plants
3D 2D
Structurefrom Motion
Leaf Segmentation Leaf Reconstruction Plant Model
![Page 21: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/21.jpg)
one of the 35 input images recovered model
Image-based Plant Modeling,Quan, Tan, Zeng, Yuan, Wang, and Kang, Siggraph 2006.
Example
Each leaf has individual geometry and texture.
![Page 22: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/22.jpg)
Structurefrom Motion
Synthesis of occluded branches
Plant ModelReconstruction ofvisible branches
by self-similarity!
But, trees have self-similar structure: a natural generative model for both leaves and branchesAll leaves are similar: a generic model for all leaves of a type of plantsThe branches are self-similar, a kind of fractal!
![Page 23: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/23.jpg)
One of the 18 input images
“This is probably the first image-based tree modeling method that produces a quality that could beadequate for movies. Earlier work produced results that were suitable for virtual reality applications, but not really anything that could be confused with a real image. Here, that seems within reach!”from one siggraph reviewer.
Recovered tree model
Image-based Tree Modeling, Tan, Zeng, Wang, Kang, and Quan, SIGGRAPH 2007.
Example
![Page 24: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/24.jpg)
Provisional summary:
generative rulesgeneric modelsData(image)-driven
![Page 25: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/25.jpg)
LargeLarge--Scale Object ModelingScale Object Modeling
![Page 26: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/26.jpg)
A traditional street in Canton:
Challenging ‘large-scale’ buildings vs ‘small-scale’ objectsSpatial constraints for capturingScale and automation
City buildings façades
![Page 27: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/27.jpg)
Generic shape: rectangle (approximation at a given scale, the ground level street-view)
Generative rules: rectangular boxes rectangles nested rectangles with different offsets
Prior knowledge on buildings and façades
![Page 28: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/28.jpg)
Basic ideas: facade by facade
Initialization as a flat rectangle in spaceDecomposition:
Dividing recursively the facade into smallrectangular patches (building blocks)Regularize the depth (quasi-dense 3D point)
Augmentation by quasi-denseCompletion (texture refetching)
Image-based Façade Modeling, Xiao, Fang, Tan, Zhao, Ofek, and Quan, SIGGRAPH ASIA 2008.
![Page 29: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/29.jpg)
![Page 30: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/30.jpg)
![Page 31: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/31.jpg)
Example for Hennepin avenue:
![Page 32: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/32.jpg)
Comparison with Virtual Earth models
![Page 33: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/33.jpg)
614 images by a handheld camera
Example of Canton street
![Page 34: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/34.jpg)
Large-scale full automation
Interactive automaticSegmentation is the keyA fully automatic approach to building modeling
![Page 35: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/35.jpg)
Classification and segmentation
Over-segmentation Segmentation (multi-view)
Classification (single view)
Input image
Five classes: buildings, sky, ground, vegetation, the others
![Page 36: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/36.jpg)
![Page 37: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/37.jpg)
Pipeline overviewPipeline overview
• (scene) classification and segmentation• (building block) decomposition scheme• (façade) structure analysis and modeling
![Page 38: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/38.jpg)
Building partition: using vertical linesBuilding partition: using vertical lines
• Detection of lines• Match only horizontal and vertical lines
using vanishing points• Reconstruct only vertical lines
for the lateral motion of the camera
![Page 39: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/39.jpg)
Building partition façadesBuilding partition façades
Decompose the buildings into small modelable blocks or facades
![Page 40: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/40.jpg)
Façade modeling: Patch reconstruction and ortho composition Façade modeling: Patch reconstruction and ortho composition
• Over-segmentation [Felzenszwalb and Huttenlocher 2004]. Estimate centroidsEstimate normal directions
• Ortho-image space (object space)
![Page 41: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/41.jpg)
Patch-reconstruction and Ortho-compositionPatch-reconstruction and Ortho-composition
From single image
From multi image
Texture mapDepth map
Input image Ortho image
![Page 42: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/42.jpg)
Façade modeling: Structure analysisFaçade modeling: Structure analysis
• Joint segmentation• Shape regularization
Segmentation Regularization
![Page 43: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/43.jpg)
Façade modeling: final resultsFaçade modeling: final results
Image-based Street-side City Modeling, Xiao, Fang, Zhao, Lhuillier, and Quan, SIGGRAPH ASIA 2009.
2.5D Façade 3D Building Texture-mapped buidling
![Page 44: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/44.jpg)
![Page 45: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/45.jpg)
Façades buildings citiesFaçades buildings cities
![Page 46: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/46.jpg)
![Page 47: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/47.jpg)
Conclusion
An image-based approach to 3DAt the intersection of vision and graphicsScale (small and large) and automation
Methodology: Generative rulesGeneric modelsData(image)-driven
![Page 48: Image-based Modelingvision.cse.psu.edu/Research/SymmComp/Workshop/Ibm-longQuan.pdfmodeling) Objects are of different natures: Smooth objects (surface reconstruction) Curvilinear objects](https://reader030.vdocument.in/reader030/viewer/2022040204/5ebfcac38b535b7f8261ca78/html5/thumbnails/48.jpg)
New bookImage-based Modeling, Long QUAN,Springer-Verlag,2010.
Part I Geometry: fundamentals of multi-view geometryPart II Computation: from pixels to 3D pointsPart III Modeling: from 3D points to objects