Computer Vision II Building Rome in a Day
|
|
|
- Shannon Gordon
- 7 years ago
- Views:
Transcription
1 Computer Vision II Building Rome in a Day
2 Recall Fundamental matrix song RANSAC song
3
4 How? Reading: the MVG bible (need ~2 years) 5-6 years Yes Coding PhD? No Crying: Not Working At All
5 + + =
6 Steps + = Images Points: Structure from Motion Points More points: Multiple View Stereo Points Meshes: Model Fitting Meshes Models: Texture Mapping Images Models: Image-based Modeling
7 Steps + = Images Points: Structure from Motion Points More points: Multiple View Stereo Points Meshes: Model Fitting Meshes Models: Texture Mapping Images Models: Image-based Modeling + + =
8 Steps + = Images Points: Structure from Motion Points More points: Multiple View Stereo Points Meshes: Model Fitting Meshes Models: Texture Mapping Images Models: Image-based Modeling + + =
9 Steps + = Images Points: Structure from Motion Points More points: Multiple View Stereo Points Meshes: Model Fitting Meshes Models: Texture Mapping Images Models: Image-based Modeling =
10 Steps + = Images Points: Structure from Motion Points More points: Multiple View Stereo Points Meshes: Model Fitting Meshes Models: Texture Mapping Images Models: Image-based Modeling
11 Steps + = Images Points: Structure from Motion Points More points: Multiple View Stereo Points Meshes: Model Fitting Meshes Models: Texture Mapping Images Models: Image-based Modeling
12 Steps + = Images Points: Structure from Motion Points More points: Multiple View Stereo Points Meshes: Model Fitting Meshes Models: Texture Mapping Images Models: Image-based Modeling
13 Image-based Modeling
14 Image-based Modeling
15 Steps + = Images Points: Structure from Motion Points More points: Multiple View Stereo Points Meshes: Model Fitting Meshes Models: Texture Mapping Images Models: Image-based Modeling + + =
16 Steps + = Images Points: Structure from Motion Points More points: Multiple View Stereo Points Meshes: Model Fitting Meshes Models: Texture Mapping Images Models: Image-based Modeling + + =
17 Steps + = Images Points: Structure from Motion Points More points: Multiple View Stereo Points Meshes: Model Fitting Meshes Models: Texture Mapping Images Models: Image-based Modeling + + =
18 Structure From Motion Structure = 3D Point Cloud of the Scene Motion = Camera Location and Orientation SFM = Get the Point Cloud from Moving Cameras Structure and Motion: Joint Problems to Solve + + =
19 Pipeline Structure from Motion (SFM) Multi-view Stereo (MVS)
20 Pipeline Structure from Motion (SFM) Multi-view Stereo (MVS)
21 Two-view Reconstruction
22 Two-view Reconstruction
23 Two-view Reconstruction keypoints keypoints match fundamental matrix essential matrix [R t] triangulation
24 Keypoints Detection keypoints keypoints match fundamental matrix essential matrix [R t] triangulation
25 Descriptor for each point SIFT descriptor SIFT descriptor keypoints keypoints match fundamental matrix essential matrix [R t] triangulation
26 Same for the other images SIFT descriptor SIFT descriptor SIFT descriptor SIFT descriptor keypoints keypoints match fundamental matrix essential matrix [R t] triangulation
27 Point Match for correspondences SIFT descriptor SIFT descriptor SIFT descriptor SIFT descriptor keypoints keypoints match fundamental matrix essential matrix [R t] triangulation
28 Point Match for correspondences SIFT descriptor SIFT descriptor SIFT descriptor SIFT descriptor keypoints keypoints match fundamental matrix essential matrix [R t] triangulation
29 Fundamental Matrix Image 1 R 1,t 1 Image 2 R 2,t 2
30 Exercise What is the difference between Fundamental Matrix and Homography? (Both of them are to explain 2D point to 2D point correspondences.)
31 Estimating Fundamental Matrix Given a correspondence The basic incidence relation is Need 8 points
32 Estimating Fundamental Matrix for 8 point correspondences: Direct Linear Transformation (DLT)
33 Algebraic Error vs. Geometric Error Algebraic Error Geometric Error (better) Unit: pixel Solved by (non-linear) least square solver (e.g. Ceres)
34 RANSAC to Estimate Fundamental Matrix For many times Pick 8 points Compute a solution for using these 8 points Count number of inliers that with close to 0 Pick the one with the largest number of inliers
35 Minimal problems in Computer Vision
36 Fundamental Matrix Essential Matrix Image 1 R 1,t 1 Image 2 R 2,t 2
37 Essential Matrix Image 1 R 1,t 1 Image 2 R 2,t 2
38 Essential Matrix Result For a given essential matrix and the first camera matrix, there are four possible choices for the second camera matrix : Page 259 of the bible (Multiple View Geometry, 2 nd Ed)
39 Four Possible Solutions Page 260 of the bible (Multiple View Geometry, 2 nd Ed)
40 Triangulation Image 1 R 1,t 1 Image 2 R 2,t 2
41 In front of the camera? Camera Extrinsic Camera Center View Direction Camera Coordinate System World Coordinate System
42 In front of the camera? A point Direction from camera center to point Angle Between Two Vectors Angle Between Just need to test and View Direction
43 Pick the Solution With maximal number of points in front of both cameras. Page 260 of the bible (Multiple View Geometry, 2 nd Ed)
44 Two-view Reconstruction keypoints keypoints match fundamental matrix essential matrix [R t] triangulation
45 Pipeline Structure from Motion (SFM) Multi-view Stereo (MVS)
46 Pipeline Taught Next
47 Merge Two Point Cloud
48 Merge Two Point Cloud There can be only one
49 Merge Two Point Cloud From the 1 st and 2 nd images, we have and From the 2 nd and 3 rd images, we have and Exercise: How to transform the coordinate system of the second point cloud to align with the first point cloud so that there is only one?
50 Merge Two Point Cloud
51 Oops See From a Different Angle
52 Bundle Adjustment
53 I am very very sexy Image 1 R 1,t 1 Image 3 R 3,t 3 Image 2 R 2,t 2
54 I am very very sexy Point 1 Point 2 Point 3 Image 1 Image 2 Image 3
55 Rethinking the SFM problem Input: Observed 2D image position Output: Unknown Camera Parameters (with some guess) Unknown Point 3D coordinate (with some guess)
56 Bundle Adjustment A valid solution must let and Re-projection = Observation
57 Bundle Adjustment A valid solution and must let the Re-projection close to the Observation, i.e. to minimize the reprojection error
58 Bundle Adjustment A valid solution and must let the Re-projection close to the Observation, i.e. to minimize the reprojection error Question: What is the unit of this objective function?
59 Camera Bundle Adjustment A valid solution and must let the Re-projection close to the Observation, i.e. to minimize the reprojection error Points
60 Camera Linking Linking SIFT Matching SIFT Matching Points
61 Solving This Optimization Problem Theory: The Levenberg Marquardt algorithm Practice: The Ceres-Solver from Google
62 Parameterizing Rotation Matrix 2D Rotation
63 3D Rotation Yaw, pitch and roll are Euler angles are
64 Axis-angle representation 3D Rotation Quaternions Avoid Gimbal Lock! Triplet Representation Recommendation! (3 dof) Not over-parameterized Rodrigues' rotation formula
65 Initialization Matters Input: Observed 2D image position Output: Unknown Camera Parameters (with some guess) Unknown Point 3D coordinate (with some guess)
66 Pipeline Taught Next
67 Multiple View Stereo State-of-the-art: PMVS: Accurate, Dense, and Robust Multi-View Stereopsis, Y Furukawa and J Ponce, Benchmark: A Comparison and Evaluation of Multi-View Stereo Reconstruction Algorithms. SM Seitz, B Curless, J Diebel, D Scharstein, R Szeliski Baseline: Multi-view stereo revisited. M Goesele, B Curless, SM Seitz
68 当前无法显示此图像 当前无法显示此图像 当前无法显示此图像 Key idea: Matching Propagation [1] Learning Two-view Stereo Matching, J Xiao, J Chen, DY Yeung, and L Quan, [2] Accurate, Dense, and Robust Multi-View Stereopsis, Y Furukawa and J Ponce, [3] Multi-view stereo revisited. M Goesele, B Curless, SM Seitz [4] Robust Dense Matching Using Local and Global Geometric Constraints, M Lhuillier & L Quan, In another context: [5] PatchMatch: A Randomized Correspondence Algorithm for Structural Image Editing, C Barnes, E Shechtman, A Finkelstein, and DB Goldman, 2009.
69 Simplest Matching Propagation We are going to learn a very simple algorithm
70 Descriptor: ZNCC (Zero-mean Normalized Cross-Correlation) Invariant to linear radiometric changes More conservative than others such as sum of absolute or square differences in uniform regions More tolerant in textured areas where noise might be important
71 Descriptor: ZNCC (Zero-mean Normalized Cross-Correlation) Invariant to linear radiometric changes More conservative than others such as sum of absolute or square differences in uniform regions More tolerant in textured areas where noise might be important
72 Seed for propagation
73 Matching Propagation Maintain a priority queue Q Initialize: Put all seeds into Q with their ZNCC values as scores For each iteration: Pop the match with best ZNCC score from Q Add new potential matches in their immediate spatial neighborhood into Q Safety: handle uniqueness, and propagate only on matchable area
74 Matchable Area the area with maximal gradience > threshold
75 Result
76 Another Example
77 Another Example
78 Another Example
79 Triangulation Image 1 R 1,t 1 Image 3 R 3,t 3 Image 2 R 2,t 2
80 Final Result
81 Colorize the Point Cloud
82 Complete Pipeline Structure from Motion (SFM) Multi-view Stereo (MVS)
83 Wait: How to get the focal length? Auto-calibration Self-Calibration and Metric Reconstruction in spite of Varying and Unknown Internal Camera Parameters, M Pollefeys, R Koch and L Van Gool, Grid Search to look for the solution with minimal reprojection error for f=min_f:max_f do everything, then obtain reprojection error after bundle adjustment Optimize for this value in bundle adjustment Camera Calibration (with checkerboard) EXIF of JPEG file recorded from digital camera Read the code of Bundler to understand how to convert EXIF into focal length value
84 Steps + = Images Points: Structure from Motion Points More points: Multiple View Stereo Points Meshes: Model Fitting Meshes Models: Texture Mapping Images Models: Image-based Modeling + + =
85 Steps + = Images Points: Structure from Motion Points More points: Multiple View Stereo Points Meshes: Model Fitting Meshes Models: Texture Mapping Images Models: Image-based Modeling =
86 Real World Applications Streetview Reconstruction and Recognition Photo Tourism Microsoft Photosynth 2d3, boujor (Matchmovers) and movies Robotics: SLAM Simultaneous Localization And Mapping
87 Steps + = Images Points: Structure from Motion Points More points: Multiple View Stereo Points Meshes: Model Fitting Meshes Models: Texture Mapping Images Models: Image-based Modeling
88 Point Cloud 3D Mesh Model Surface Reconstruction Marching Cubes: Poisson Surface Reconstruction Model Fitting RANSAC & J-linkage InverseCSG GlobFit Face Model Fitting
89 Maps 92
90 Photorealistic Maps 93
91 What about indoors? The Birth of Venus 94
92 Uffizi Museum 95
93 Cartography 2D Line Drawing 3D Realistic Maps Outdoor Indoor? 96 Reconstructing the World's Museums. Xiao and Furukawa, 2012.
94 Photorealistic Indoor Maps 97
95 Photorealistic Indoor Maps Where is The Birth of Venus? 98
96 Visualize & Explore Big Data 100
97 Data-driven Brute-force 2-step algorithm 102
98 Data-driven Brute-force 1. Remove ceiling. 2. Take pictures from aerial viewpoints. 103
99 What is Wrong? I don t own the museums. 104
100 Old Approach 1. Remove ceiling. 2. Take pictures from aerial viewpoints. 105
101 New Approach 0. Own the museums. 1. Remove ceiling. 2. Take pictures from aerial viewpoints. 106
102 New Approach 0. Reconstruct the museums. 1. Remove ceiling. 2. Take pictures from aerial viewpoints. 107
103 108
104 InverseCSG Algorithm Reconstructing the World's Museums, J. Xiao and Y. Furukawa, 2012.
105 InverseCSG Algorithm
106 Noisy Points top-down view of input points
107 InverseCSG Algorithm Constructive Solid Geometry (CSG)
108 InverseCSG Algorithm Constructive Solid Geometry (CSG) Cuboids as primitives Xiao et al. NIPS 2012 Xiao et al. Siggraph Asia 2012a
109 Bottom-up & Top-down Model 3D Cuboid 2D Rectangle 2D Line Point 115
110 InverseCSG Algorithm gravity side view 3D Point Cloud 2D CSG 3D CSG Wall Final 116
111 Cut into Slices side view point count 3D Point Cloud 2D CSG 3D CSG Wall Final 117
112 Cut into Slices side view 3D Point Cloud 2D CSG 3D CSG Wall Final 118
113 2D CSG Reconstruction 1. Generate primitives 2. Choose a subset 3D Point Cloud 2D CSG 3D CSG Wall Final top view 119
114 2D CSG Reconstruction 1. Generate primitives Point Line 3D Point Cloud 2D CSG 3D CSG Wall Final top view 120
115 2D CSG Reconstruction 1. Generate primitives From 4 line segments Point Line Line Rectangle 3D Point Cloud 2D CSG 3D CSG Wall Final 121
116 2D CSG Reconstruction 1. Generate primitives 2. Choose a subset 3D Point Cloud 2D CSG 3D CSG Wall Final top view 122
117 2D CSG Reconstruction 1. Generate primitives 2. Choose a subset Repeat in each slice 3D Point Cloud 2D CSG 3D CSG Wall Final top view 123
118 Objective Function Likelihood Points Free space Prior Simple Points Free space Prior top view 3D Point Cloud 2D CSG 3D CSG Wall Final 124
119 Objective Function Likelihood Points Free space Prior Simple Points Free space Prior top view 3D Point Cloud 2D CSG 3D CSG Wall Final 125
120 Objective Function Likelihood Points Free space Prior Simple Points Free space Prior top view 3D Point Cloud 2D CSG 3D CSG Wall Final 126
121 Objective Function Likelihood Points Free space Prior Simple Points Free space Prior top view 3D Point Cloud 2D CSG 3D CSG Wall Final 127
122 Objective Function Likelihood Points Free space Prior Simple Points Free space Prior top view 3D Point Cloud 2D CSG 3D CSG Wall Final 128
123 Objective Function Likelihood Points Free space Prior Simple Points Free space Prior top view 3D Point Cloud 2D CSG 3D CSG Wall Final 129
124 3D CSG Reconstruction 1. Generate primitives (cuboids) 2. Choose a subset (of primitive candidates) 3D Point Cloud 2D CSG 3D CSG Wall Final 130
125 3D CSG Reconstruction 1. Generate primitives (cuboids) side view gravity 3D Point Cloud 2D CSG 3D CSG Wall Final 131
126 3D CSG Reconstruction 1. Generate primitives (cuboids) 2D CSG side view 3D Point Cloud 2D CSG 3D CSG Wall Final 132
127 oblique view top view 2D CSG side view 3D Point Cloud 2D CSG 3D CSG Wall Final 133
128 3D CSG Reconstruction 1. Generate primitives (cuboids) Rectangle Primitive 2D CSG side view 3D Point Cloud 2D CSG 3D CSG Wall Final 134
129 3D CSG Reconstruction 1. Generate primitives (cuboids) Rectangle Primitive 2D CSG side view 3D Point Cloud 2D CSG 3D CSG Wall Final 135
130 3D CSG Reconstruction 1. Generate primitives (cuboids) 2. Choose a subset side view 3D Point Cloud 2D CSG 3D CSG Wall Final 136
131 Algorithm on Run
132 Result
133 Steps + = Images Points: Structure from Motion Points More points: Multiple View Stereo Points Meshes: Model Fitting Meshes Models: Texture Mapping Images Models: Image-based Modeling
134 Texture Mapping Texture Stitching View-based Texture Mapping View Interpolation and Warping Interactive Visualization
135 Image Warping: Forward Warping Source Image Target Image Matlab interp2 example:
136 Image Warping: Backward Warping Source Image Target Image Matlab interp2 example:
137 Texture Stitching The process of assembling projected images to form a composite rendering Image from Paul Debevec s Thesis
138 Texture Stitching Image-based Street-side City Modeling. Xiao et al, 2009.
139 Texture Stitching Image 1 R 1,t 1 Image 3 R 3,t 3 Image 2 R 2,t 2
140 Texture Stitching Image-based Street-side City Modeling. Xiao et al, 2009.
141
142
143 View-Dependent Texture Mapping Modeling and Rendering Architecture from Photographs, Paul Debevec, PhD Thesis, UC Berkeley, 1996
144 View Interpolation Match Propagation from Image-based Modeling and Rendering M Lhuillier and L Quan, 2012
145 View Interpolation
146 View Interpolation Match Propagation from Image-based Modeling and Rendering, M Lhuillier and L Quan, 2012
147 Interactive Visualization Reconstructing Building Interiors from Images, Y Furukawa, B Curless, SM Seitz, R Szeliski, 2009 Download the Viewer from:
148 Aerial+Ground Visualization Reconstructing the World's Museums. J. Xiao and Y. Furukawa,
149 Ground vs. Aerial Ground + Aerial Ground Aerial Ground+Aerial Outdoor Google Streetview Google/Bing/NASA Google MapsGL Indoor Furukawa et al. Xiao et al. Xiao et al.
150 Finish our Journey! + = Images Points: Structure from Motion Points More points: Multiple View Stereo Points Meshes: Model Fitting Meshes Models: Texture Mapping Images Models: Image-based Modeling + + =
151 History of 3D Reconstruction
152 Longer Summary of the History Research Landmarks for 3D Reconstruction Steve Seitz "History of 3D Computer Vision NSF Frontiers in Computer Vision,
Build Panoramas on Android Phones
Build Panoramas on Android Phones Tao Chu, Bowen Meng, Zixuan Wang Stanford University, Stanford CA Abstract The purpose of this work is to implement panorama stitching from a sequence of photos taken
Interactive Dense 3D Modeling of Indoor Environments
Interactive Dense 3D Modeling of Indoor Environments Hao Du 1 Peter Henry 1 Xiaofeng Ren 2 Dieter Fox 1,2 Dan B Goldman 3 Steven M. Seitz 1 {duhao,peter,fox,seitz}@cs.washinton.edu xiaofeng.ren@intel.com
CS 534: Computer Vision 3D Model-based recognition
CS 534: Computer Vision 3D Model-based recognition Ahmed Elgammal Dept of Computer Science CS 534 3D Model-based Vision - 1 High Level Vision Object Recognition: What it means? Two main recognition tasks:!
Two hours UNIVERSITY OF MANCHESTER SCHOOL OF COMPUTER SCIENCE. M.Sc. in Advanced Computer Science. Friday 18 th January 2008.
COMP60321 Two hours UNIVERSITY OF MANCHESTER SCHOOL OF COMPUTER SCIENCE M.Sc. in Advanced Computer Science Computer Animation Friday 18 th January 2008 Time: 09:45 11:45 Please answer any THREE Questions
Image Processing and Computer Graphics. Rendering Pipeline. Matthias Teschner. Computer Science Department University of Freiburg
Image Processing and Computer Graphics Rendering Pipeline Matthias Teschner Computer Science Department University of Freiburg Outline introduction rendering pipeline vertex processing primitive processing
3D Tranformations. CS 4620 Lecture 6. Cornell CS4620 Fall 2013 Lecture 6. 2013 Steve Marschner (with previous instructors James/Bala)
3D Tranformations CS 4620 Lecture 6 1 Translation 2 Translation 2 Translation 2 Translation 2 Scaling 3 Scaling 3 Scaling 3 Scaling 3 Rotation about z axis 4 Rotation about z axis 4 Rotation about x axis
3D Modeling and Simulation using Image Stitching
3D Modeling and Simulation using Image Stitching Sean N. Braganza K. J. Somaiya College of Engineering, Mumbai, India ShubhamR.Langer K. J. Somaiya College of Engineering,Mumbai, India Pallavi G.Bhoite
Feature Tracking and Optical Flow
02/09/12 Feature Tracking and Optical Flow Computer Vision CS 543 / ECE 549 University of Illinois Derek Hoiem Many slides adapted from Lana Lazebnik, Silvio Saverse, who in turn adapted slides from Steve
MetropoGIS: A City Modeling System DI Dr. Konrad KARNER, DI Andreas KLAUS, DI Joachim BAUER, DI Christopher ZACH
MetropoGIS: A City Modeling System DI Dr. Konrad KARNER, DI Andreas KLAUS, DI Joachim BAUER, DI Christopher ZACH VRVis Research Center for Virtual Reality and Visualization, Virtual Habitat, Inffeldgasse
Metrics on SO(3) and Inverse Kinematics
Mathematical Foundations of Computer Graphics and Vision Metrics on SO(3) and Inverse Kinematics Luca Ballan Institute of Visual Computing Optimization on Manifolds Descent approach d is a ascent direction
Epipolar Geometry. Readings: See Sections 10.1 and 15.6 of Forsyth and Ponce. Right Image. Left Image. e(p ) Epipolar Lines. e(q ) q R.
Epipolar Geometry We consider two perspective images of a scene as taken from a stereo pair of cameras (or equivalently, assume the scene is rigid and imaged with a single camera from two different locations).
Introduction to Computer Graphics
Introduction to Computer Graphics Torsten Möller TASC 8021 778-782-2215 torsten@sfu.ca www.cs.sfu.ca/~torsten Today What is computer graphics? Contents of this course Syllabus Overview of course topics
3D Model based Object Class Detection in An Arbitrary View
3D Model based Object Class Detection in An Arbitrary View Pingkun Yan, Saad M. Khan, Mubarak Shah School of Electrical Engineering and Computer Science University of Central Florida http://www.eecs.ucf.edu/
Bildverarbeitung und Mustererkennung Image Processing and Pattern Recognition
Bildverarbeitung und Mustererkennung Image Processing and Pattern Recognition 1. Image Pre-Processing - Pixel Brightness Transformation - Geometric Transformation - Image Denoising 1 1. Image Pre-Processing
CIS 536/636 Introduction to Computer Graphics. Kansas State University. CIS 536/636 Introduction to Computer Graphics
2 Lecture Outline Animation 2 of 3: Rotations, Quaternions Dynamics & Kinematics William H. Hsu Department of Computing and Information Sciences, KSU KSOL course pages: http://bit.ly/hgvxlh / http://bit.ly/evizre
Classifying Manipulation Primitives from Visual Data
Classifying Manipulation Primitives from Visual Data Sandy Huang and Dylan Hadfield-Menell Abstract One approach to learning from demonstrations in robotics is to make use of a classifier to predict if
On Benchmarking Camera Calibration and Multi-View Stereo for High Resolution Imagery
On Benchmarking Camera Calibration and Multi-View Stereo for High Resolution Imagery C. Strecha CVLab EPFL Lausanne (CH) W. von Hansen FGAN-FOM Ettlingen (D) L. Van Gool CVLab ETHZ Zürich (CH) P. Fua CVLab
ACCURACY ASSESSMENT OF BUILDING POINT CLOUDS AUTOMATICALLY GENERATED FROM IPHONE IMAGES
ACCURACY ASSESSMENT OF BUILDING POINT CLOUDS AUTOMATICALLY GENERATED FROM IPHONE IMAGES B. Sirmacek, R. Lindenbergh Delft University of Technology, Department of Geoscience and Remote Sensing, Stevinweg
Essential Mathematics for Computer Graphics fast
John Vince Essential Mathematics for Computer Graphics fast Springer Contents 1. MATHEMATICS 1 Is mathematics difficult? 3 Who should read this book? 4 Aims and objectives of this book 4 Assumptions made
Introduction to Computer Graphics Marie-Paule Cani & Estelle Duveau
Introduction to Computer Graphics Marie-Paule Cani & Estelle Duveau 04/02 Introduction & projective rendering 11/02 Prodedural modeling, Interactive modeling with parametric surfaces 25/02 Introduction
3D Scanner using Line Laser. 1. Introduction. 2. Theory
. Introduction 3D Scanner using Line Laser Di Lu Electrical, Computer, and Systems Engineering Rensselaer Polytechnic Institute The goal of 3D reconstruction is to recover the 3D properties of a geometric
3 Image-Based Photo Hulls. 2 Image-Based Visual Hulls. 3.1 Approach. 3.2 Photo-Consistency. Figure 1. View-dependent geometry.
Image-Based Photo Hulls Greg Slabaugh, Ron Schafer Georgia Institute of Technology Center for Signal and Image Processing Atlanta, GA 30332 {slabaugh, rws}@ece.gatech.edu Mat Hans Hewlett-Packard Laboratories
Part-Based Recognition
Part-Based Recognition Benedict Brown CS597D, Fall 2003 Princeton University CS 597D, Part-Based Recognition p. 1/32 Introduction Many objects are made up of parts It s presumably easier to identify simple
Real-Time Stereo Reconstruction in Robotically Assisted Minimally Invasive Surgery
Real-Time Stereo Reconstruction in Robotically Assisted Minimally Invasive Surgery Abstract. The recovery of tissue structure and morphology during robotic assisted surgery is an important step towards
Building an Advanced Invariant Real-Time Human Tracking System
UDC 004.41 Building an Advanced Invariant Real-Time Human Tracking System Fayez Idris 1, Mazen Abu_Zaher 2, Rashad J. Rasras 3, and Ibrahiem M. M. El Emary 4 1 School of Informatics and Computing, German-Jordanian
Robust NURBS Surface Fitting from Unorganized 3D Point Clouds for Infrastructure As-Built Modeling
81 Robust NURBS Surface Fitting from Unorganized 3D Point Clouds for Infrastructure As-Built Modeling Andrey Dimitrov 1 and Mani Golparvar-Fard 2 1 Graduate Student, Depts of Civil Eng and Engineering
ENGN 2502 3D Photography / Winter 2012 / SYLLABUS http://mesh.brown.edu/3dp/
ENGN 2502 3D Photography / Winter 2012 / SYLLABUS http://mesh.brown.edu/3dp/ Description of the proposed course Over the last decade digital photography has entered the mainstream with inexpensive, miniaturized
Environmental Remote Sensing GEOG 2021
Environmental Remote Sensing GEOG 2021 Lecture 4 Image classification 2 Purpose categorising data data abstraction / simplification data interpretation mapping for land cover mapping use land cover class
MeshLab and Arc3D: Photo-Reconstruction and Processing of 3D meshes
MeshLab and Arc3D: Photo-Reconstruction and Processing of 3D meshes P. Cignoni, M Corsini, M. Dellepiane, G. Ranzuglia, (Visual Computing Lab, ISTI - CNR, Italy) M. Vergauven, L. Van Gool (K.U.Leuven ESAT-PSI
Probabilistic Latent Semantic Analysis (plsa)
Probabilistic Latent Semantic Analysis (plsa) SS 2008 Bayesian Networks Multimedia Computing, Universität Augsburg Rainer.Lienhart@informatik.uni-augsburg.de www.multimedia-computing.{de,org} References
Image Segmentation and Registration
Image Segmentation and Registration Dr. Christine Tanner (tanner@vision.ee.ethz.ch) Computer Vision Laboratory, ETH Zürich Dr. Verena Kaynig, Machine Learning Laboratory, ETH Zürich Outline Segmentation
OmniTour: Semi-automatic generation of interactive virtual tours from omnidirectional video
OmniTour: Semi-automatic generation of interactive virtual tours from omnidirectional video Olivier Saurer Friedrich Fraundorfer Marc Pollefeys Computer Vision and Geometry Group ETH Zürich, Switzerland
Wii Remote Calibration Using the Sensor Bar
Wii Remote Calibration Using the Sensor Bar Alparslan Yildiz Abdullah Akay Yusuf Sinan Akgul GIT Vision Lab - http://vision.gyte.edu.tr Gebze Institute of Technology Kocaeli, Turkey {yildiz, akay, akgul}@bilmuh.gyte.edu.tr
A Study on SURF Algorithm and Real-Time Tracking Objects Using Optical Flow
, pp.233-237 http://dx.doi.org/10.14257/astl.2014.51.53 A Study on SURF Algorithm and Real-Time Tracking Objects Using Optical Flow Giwoo Kim 1, Hye-Youn Lim 1 and Dae-Seong Kang 1, 1 Department of electronices
RGB-D Mapping: Using Kinect-Style Depth Cameras for Dense 3D Modeling of Indoor Environments
RGB-D Mapping: Using Kinect-Style Depth Cameras for Dense 3D Modeling of Indoor Environments Peter Henry 1, Michael Krainin 1, Evan Herbst 1, Xiaofeng Ren 2, Dieter Fox 1 Abstract RGB-D cameras (such as
Segmentation of building models from dense 3D point-clouds
Segmentation of building models from dense 3D point-clouds Joachim Bauer, Konrad Karner, Konrad Schindler, Andreas Klaus, Christopher Zach VRVis Research Center for Virtual Reality and Visualization, Institute
THE PERFORMANCE EVALUATION OF MULTI-IMAGE 3D RECONSTRUCTION SOFTWARE WITH DIFFERENT SENSORS
International Conference on Sensors & Models in Remote Sensing & Photogrammetry, 23 25 v 2015, Kish Island, Iran THE PERFORMANCE EVALUATION OF MULTI-IMAGE 3D RECONSTRUCTION SOFTWARE WITH DIFFERENT SENSORS
State of the Art and Challenges in Crowd Sourced Modeling
Photogrammetric Week '11 Dieter Fritsch (Ed.) Wichmann/VDE Verlag, Belin & Offenbach, 2011 Frahm et al. 259 State of the Art and Challenges in Crowd Sourced Modeling JAN-MICHAEL FRAHM, PIERRE FITE-GEORGEL,
INTRODUCTION TO RENDERING TECHNIQUES
INTRODUCTION TO RENDERING TECHNIQUES 22 Mar. 212 Yanir Kleiman What is 3D Graphics? Why 3D? Draw one frame at a time Model only once X 24 frames per second Color / texture only once 15, frames for a feature
Computer Graphics. Geometric Modeling. Page 1. Copyright Gotsman, Elber, Barequet, Karni, Sheffer Computer Science - Technion. An Example.
An Example 2 3 4 Outline Objective: Develop methods and algorithms to mathematically model shape of real world objects Categories: Wire-Frame Representation Object is represented as as a set of points
Point Cloud Simulation & Applications Maurice Fallon
Point Cloud & Applications Maurice Fallon Contributors: MIT: Hordur Johannsson and John Leonard U. of Salzburg: Michael Gschwandtner and Roland Kwitt Overview : Dense disparity information Efficient Image
IMPLICIT SHAPE MODELS FOR OBJECT DETECTION IN 3D POINT CLOUDS
IMPLICIT SHAPE MODELS FOR OBJECT DETECTION IN 3D POINT CLOUDS Alexander Velizhev 1 (presenter) Roman Shapovalov 2 Konrad Schindler 3 1 Hexagon Technology Center, Heerbrugg, Switzerland 2 Graphics & Media
How To Fuse A Point Cloud With A Laser And Image Data From A Pointcloud
REAL TIME 3D FUSION OF IMAGERY AND MOBILE LIDAR Paul Mrstik, Vice President Technology Kresimir Kusevic, R&D Engineer Terrapoint Inc. 140-1 Antares Dr. Ottawa, Ontario K2E 8C4 Canada paul.mrstik@terrapoint.com
Removing Moving Objects from Point Cloud Scenes
1 Removing Moving Objects from Point Cloud Scenes Krystof Litomisky klitomis@cs.ucr.edu Abstract. Three-dimensional simultaneous localization and mapping is a topic of significant interest in the research
Terrain Traversability Analysis using Organized Point Cloud, Superpixel Surface Normals-based segmentation and PCA-based Classification
Terrain Traversability Analysis using Organized Point Cloud, Superpixel Surface Normals-based segmentation and PCA-based Classification Aras Dargazany 1 and Karsten Berns 2 Abstract In this paper, an stereo-based
Spatio-Temporally Coherent 3D Animation Reconstruction from Multi-view RGB-D Images using Landmark Sampling
, March 13-15, 2013, Hong Kong Spatio-Temporally Coherent 3D Animation Reconstruction from Multi-view RGB-D Images using Landmark Sampling Naveed Ahmed Abstract We present a system for spatio-temporally
Photo Tourism: Exploring Photo Collections in 3D
Photo Tourism: Exploring Photo Collections in 3D (a) (b) (c) Figure 1: Our system takes unstructured collections of photographs such as those from online image searches (a) and reconstructs 3D points and
Incremental Surface Extraction from Sparse Structure-from-Motion Point Clouds
HOPPE, KLOPSCHITZ, DONOSER, BISCHOF: INCREMENTAL SURFACE EXTRACTION 1 Incremental Surface Extraction from Sparse Structure-from-Motion Point Clouds Christof Hoppe 1 hoppe@icg.tugraz.at Manfred Klopschitz
How To Analyze Ball Blur On A Ball Image
Single Image 3D Reconstruction of Ball Motion and Spin From Motion Blur An Experiment in Motion from Blur Giacomo Boracchi, Vincenzo Caglioti, Alessandro Giusti Objective From a single image, reconstruct:
EECS 556 Image Processing W 09. Interpolation. Interpolation techniques B splines
EECS 556 Image Processing W 09 Interpolation Interpolation techniques B splines What is image processing? Image processing is the application of 2D signal processing methods to images Image representation
Virtual Zoom: Augmented Reality on a Mobile Device
Virtual Zoom: Augmented Reality on a Mobile Device Sergey Karayev University of Washington Honors Thesis in Computer Science June 5, 2009 Abstract The world around us is photographed millions of times
Jiří Matas. Hough Transform
Hough Transform Jiří Matas Center for Machine Perception Department of Cybernetics, Faculty of Electrical Engineering Czech Technical University, Prague Many slides thanks to Kristen Grauman and Bastian
Object Recognition and Template Matching
Object Recognition and Template Matching Template Matching A template is a small image (sub-image) The goal is to find occurrences of this template in a larger image That is, you want to find matches of
CHAPTER 6 TEXTURE ANIMATION
CHAPTER 6 TEXTURE ANIMATION 6.1. INTRODUCTION Animation is the creating of a timed sequence or series of graphic images or frames together to give the appearance of continuous movement. A collection of
Assessment. Presenter: Yupu Zhang, Guoliang Jin, Tuo Wang Computer Vision 2008 Fall
Automatic Photo Quality Assessment Presenter: Yupu Zhang, Guoliang Jin, Tuo Wang Computer Vision 2008 Fall Estimating i the photorealism of images: Distinguishing i i paintings from photographs h Florin
Using Photorealistic RenderMan for High-Quality Direct Volume Rendering
Using Photorealistic RenderMan for High-Quality Direct Volume Rendering Cyrus Jam cjam@sdsc.edu Mike Bailey mjb@sdsc.edu San Diego Supercomputer Center University of California San Diego Abstract With
Colorado School of Mines Computer Vision Professor William Hoff
Professor William Hoff Dept of Electrical Engineering &Computer Science http://inside.mines.edu/~whoff/ 1 Introduction to 2 What is? A process that produces from images of the external world a description
Dense Matching Methods for 3D Scene Reconstruction from Wide Baseline Images
Dense Matching Methods for 3D Scene Reconstruction from Wide Baseline Images Zoltán Megyesi PhD Theses Supervisor: Prof. Dmitry Chetverikov Eötvös Loránd University PhD Program in Informatics Program Director:
EXPERIMENTAL EVALUATION OF RELATIVE POSE ESTIMATION ALGORITHMS
EXPERIMENTAL EVALUATION OF RELATIVE POSE ESTIMATION ALGORITHMS Marcel Brückner, Ferid Bajramovic, Joachim Denzler Chair for Computer Vision, Friedrich-Schiller-University Jena, Ernst-Abbe-Platz, 7743 Jena,
Files Used in this Tutorial
Generate Point Clouds Tutorial This tutorial shows how to generate point clouds from IKONOS satellite stereo imagery. You will view the point clouds in the ENVI LiDAR Viewer. The estimated time to complete
3D Modelling in Blender Based on Polygonal Data
3D Modelling in Blender Based on Polygonal Data James Ribe MSCS Department St. Olaf College 1500 St. Olaf Ave Northfield, MN 55438 ribe@stolaf.edu Alora Killian MSCS Department St. Olaf College 1500 St.
animation animation shape specification as a function of time
animation animation shape specification as a function of time animation representation many ways to represent changes with time intent artistic motion physically-plausible motion efficiency control typically
3D Object Recognition in Clutter with the Point Cloud Library
3D Object Recognition in Clutter with the Point Cloud Library Federico Tombari, Ph.D federico.tombari@unibo.it University of Bologna Open Perception Data representations in PCL PCL can deal with both organized
PHOTOGRAMMETRIC MODELING AND IMAGE-BASED RENDERING FOR RAPID VIRTUAL ENVIRONMENT CREATION
PHOTOGRAMMETRIC MODELING AND IMAGE-BASED RENDERING FOR RAPID VIRTUAL ENVIRONMENT CREATION Charalambos Poullis*, Andrew Gardner, Paul Debevec University of Southern California Institute for Creative Technologies
Modelling 3D Avatar for Virtual Try on
Modelling 3D Avatar for Virtual Try on NADIA MAGNENAT THALMANN DIRECTOR MIRALAB UNIVERSITY OF GENEVA DIRECTOR INSTITUTE FOR MEDIA INNOVATION, NTU, SINGAPORE WWW.MIRALAB.CH/ Creating Digital Humans Vertex
Photo Uncrop. 1 Introduction. Qi Shan, Brian Curless, Yasutaka Furukawa, Carlos Hernandez, and Steven M. Seitz
Photo Uncrop Qi Shan, Brian Curless, Yasutaka Furukawa, Carlos Hernandez, and Steven M. Seitz University of Washington Washington University in St. Louis Google Abstract. We address the problem of extending
Bachelor of Games and Virtual Worlds (Programming) Subject and Course Summaries
First Semester Development 1A On completion of this subject students will be able to apply basic programming and problem solving skills in a 3 rd generation object-oriented programming language (such as
Camera geometry and image alignment
Computer Vision and Machine Learning Winter School ENS Lyon 2010 Camera geometry and image alignment Josef Sivic http://www.di.ens.fr/~josef INRIA, WILLOW, ENS/INRIA/CNRS UMR 8548 Laboratoire d Informatique,
A typical 3D modeling system involves the phases of 1. Individual range image acquisition from different viewpoints.
Efficient Model Creation of Large Structures based on Range Segmentation 2nd International Symposium on 3D Data Processing, Visualization & Transmission, September 2004, Thessaloniki, Greece. Ioannis Stamos
GUIDE TO POST-PROCESSING OF THE POINT CLOUD
GUIDE TO POST-PROCESSING OF THE POINT CLOUD Contents Contents 3 Reconstructing the point cloud with MeshLab 16 Reconstructing the point cloud with CloudCompare 2 Reconstructing the point cloud with MeshLab
Photo Tourism: Exploring Photo Collections in 3D
Photo Tourism: Exploring Photo Collections in 3D Noah Snavely University of Washington Steven M. Seitz University of Washington Richard Szeliski Microsoft Research (a) (b) (c) Figure 1: Our system takes
Kinematical Animation. lionel.reveret@inria.fr 2013-14
Kinematical Animation 2013-14 3D animation in CG Goal : capture visual attention Motion of characters Believable Expressive Realism? Controllability Limits of purely physical simulation : - little interactivity
A Prototype For Eye-Gaze Corrected
A Prototype For Eye-Gaze Corrected Video Chat on Graphics Hardware Maarten Dumont, Steven Maesen, Sammy Rogmans and Philippe Bekaert Introduction Traditional webcam video chat: No eye contact. No extensive
Towards Linear-time Incremental Structure from Motion
Towards Linear-time Incremental Structure from Motion Changchang Wu University of Washington Abstract The time complexity of incremental structure from motion (SfM) is often known as O(n 4 ) with respect
Computer Animation. Lecture 2. Basics of Character Animation
Computer Animation Lecture 2. Basics of Character Animation Taku Komura Overview Character Animation Posture representation Hierarchical structure of the body Joint types Translational, hinge, universal,
Randomized Trees for Real-Time Keypoint Recognition
Randomized Trees for Real-Time Keypoint Recognition Vincent Lepetit Pascal Lagger Pascal Fua Computer Vision Laboratory École Polytechnique Fédérale de Lausanne (EPFL) 1015 Lausanne, Switzerland Email:
Optical Tracking Using Projective Invariant Marker Pattern Properties
Optical Tracking Using Projective Invariant Marker Pattern Properties Robert van Liere, Jurriaan D. Mulder Department of Information Systems Center for Mathematics and Computer Science Amsterdam, the Netherlands
Introduction Epipolar Geometry Calibration Methods Further Readings. Stereo Camera Calibration
Stereo Camera Calibration Stereo Camera Calibration Stereo Camera Calibration Stereo Camera Calibration 12.10.2004 Overview Introduction Summary / Motivation Depth Perception Ambiguity of Correspondence
Are we ready for Autonomous Driving? The KITTI Vision Benchmark Suite
Are we ready for Autonomous Driving? The KITTI Vision Benchmark Suite Philip Lenz 1 Andreas Geiger 2 Christoph Stiller 1 Raquel Urtasun 3 1 KARLSRUHE INSTITUTE OF TECHNOLOGY 2 MAX-PLANCK-INSTITUTE IS 3
A Genetic Algorithm-Evolved 3D Point Cloud Descriptor
A Genetic Algorithm-Evolved 3D Point Cloud Descriptor Dominik Wȩgrzyn and Luís A. Alexandre IT - Instituto de Telecomunicações Dept. of Computer Science, Univ. Beira Interior, 6200-001 Covilhã, Portugal
Monocular Model-Based 3D Tracking of Rigid Objects: A Survey
Foundations and Trends R in Computer Graphics and Vision Vol. 1, No 1 (2005) 1 89 c 2005 V. Lepetit and P. Fua Monocular Model-Based 3D Tracking of Rigid Objects: A Survey Vincent Lepetit 1 and Pascal
BUILDING TELEPRESENCE SYSTEMS: Translating Science Fiction Ideas into Reality
BUILDING TELEPRESENCE SYSTEMS: Translating Science Fiction Ideas into Reality Henry Fuchs University of North Carolina at Chapel Hill (USA) and NSF Science and Technology Center for Computer Graphics and
Automatic georeferencing of imagery from high-resolution, low-altitude, low-cost aerial platforms
Automatic georeferencing of imagery from high-resolution, low-altitude, low-cost aerial platforms Amanda Geniviva, Jason Faulring and Carl Salvaggio Rochester Institute of Technology, 54 Lomb Memorial
CATIA V5 Tutorials. Mechanism Design & Animation. Release 18. Nader G. Zamani. University of Windsor. Jonathan M. Weaver. University of Detroit Mercy
CATIA V5 Tutorials Mechanism Design & Animation Release 18 Nader G. Zamani University of Windsor Jonathan M. Weaver University of Detroit Mercy SDC PUBLICATIONS Schroff Development Corporation www.schroff.com
Online Environment Mapping
Online Environment Mapping Jongwoo Lim Honda Research Institute USA, inc. Mountain View, CA, USA jlim@honda-ri.com Jan-Michael Frahm Department of Computer Science University of North Carolina Chapel Hill,
Auto Head-Up Displays: View-Through for Drivers
Auto Head-Up Displays: View-Through for Drivers Head-up displays (HUD) are featuring more and more frequently in both current and future generation automobiles. One primary reason for this upward trend
PHYSIOLOGICALLY-BASED DETECTION OF COMPUTER GENERATED FACES IN VIDEO
PHYSIOLOGICALLY-BASED DETECTION OF COMPUTER GENERATED FACES IN VIDEO V. Conotter, E. Bodnari, G. Boato H. Farid Department of Information Engineering and Computer Science University of Trento, Trento (ITALY)
PCL - SURFACE RECONSTRUCTION
PCL - SURFACE RECONSTRUCTION TOYOTA CODE SPRINT Alexandru-Eugen Ichim Computer Graphics and Geometry Laboratory PROBLEM DESCRIPTION 1/2 3D revolution due to cheap RGB-D cameras (Asus Xtion & Microsoft
Direct Volume Rendering Elvins
Direct Volume Rendering Elvins 1 Principle: rendering of scalar volume data with cloud-like, semi-transparent effects forward mapping object-order backward mapping image-order data volume screen voxel
3D U ser I t er aces and Augmented Reality
3D User Interfaces and Augmented Reality Applications Mechanical CAD 3D Animation Virtual Environments Scientific Visualization Mechanical CAD Component design Assembly testingti Mechanical properties
Automatic 3D Reconstruction via Object Detection and 3D Transformable Model Matching CS 269 Class Project Report
Automatic 3D Reconstruction via Object Detection and 3D Transformable Model Matching CS 69 Class Project Report Junhua Mao and Lunbo Xu University of California, Los Angeles mjhustc@ucla.edu and lunbo
Consolidated Visualization of Enormous 3D Scan Point Clouds with Scanopy
Consolidated Visualization of Enormous 3D Scan Point Clouds with Scanopy Claus SCHEIBLAUER 1 / Michael PREGESBAUER 2 1 Institute of Computer Graphics and Algorithms, Vienna University of Technology, Austria
Edge-based Template Matching and Tracking for Perspectively Distorted Planar Objects
Edge-based Template Matching and Tracking for Perspectively Distorted Planar Objects Andreas Hofhauser, Carsten Steger, and Nassir Navab TU München, Boltzmannstr. 3, 85748 Garching bei München, Germany
Automotive Applications of 3D Laser Scanning Introduction
Automotive Applications of 3D Laser Scanning Kyle Johnston, Ph.D., Metron Systems, Inc. 34935 SE Douglas Street, Suite 110, Snoqualmie, WA 98065 425-396-5577, www.metronsys.com 2002 Metron Systems, Inc
Fast and Robust Normal Estimation for Point Clouds with Sharp Features
1/37 Fast and Robust Normal Estimation for Point Clouds with Sharp Features Alexandre Boulch & Renaud Marlet University Paris-Est, LIGM (UMR CNRS), Ecole des Ponts ParisTech Symposium on Geometry Processing
GPS-aided Recognition-based User Tracking System with Augmented Reality in Extreme Large-scale Areas
GPS-aided Recognition-based User Tracking System with Augmented Reality in Extreme Large-scale Areas Wei Guan Computer Graphics and Immersive Technologies Computer Science, USC wguan@usc.edu Suya You Computer
ConstructAide: Analyzing and Visualizing Construction Sites through Photographs and Building Models
ConstructAide: Analyzing and Visualizing Construction Sites through Photographs and Building Models Kevin Karsch Mani Golparvar-Fard David Forsyth University of Illinois Future Behind Schedule Ahead of
An introduction to 3D draughting & solid modelling using AutoCAD
An introduction to 3D draughting & solid modelling using AutoCAD Faculty of Technology University of Plymouth Drake Circus Plymouth PL4 8AA These notes are to be used in conjunction with the AutoCAD software
Space Perception and Binocular Vision
Space Perception and Binocular Vision Space Perception Monocular Cues to Three-Dimensional Space Binocular Vision and Stereopsis Combining Depth Cues 9/30/2008 1 Introduction to Space Perception Realism:
A Short Introduction to Computer Graphics
A Short Introduction to Computer Graphics Frédo Durand MIT Laboratory for Computer Science 1 Introduction Chapter I: Basics Although computer graphics is a vast field that encompasses almost any graphical