Niloy J. Mitra Leonidas J. Guibas Mark Pauly TU Vienna Stanford University ETH Zurich SIGGRAPH 2007.

Slides:



Advertisements
Similar presentations
Coherent Laplacian 3D protrusion segmentation Oxford Brookes Vision Group Queen Mary, University of London, 11/12/2009 Fabio Cuzzolin.
Advertisements

Partial and Approximate Symmetry Detection for 3D Geometry Mark Pauly Niloy J. Mitra Leonidas J. Guibas.
Robust Global Registration Natasha Gelfand Niloy Mitra Leonidas Guibas Helmut Pottmann.
Two-View Geometry CS Sastry and Yang
Probabilistic Fingerprints for Shapes Niloy J. MitraLeonidas Guibas Joachim GiesenMark Pauly Stanford University MPII SaarbrückenETH Zurich.
Instructor: Mircea Nicolescu Lecture 13 CS 485 / 685 Computer Vision.
Relevance Feedback Content-Based Image Retrieval Using Query Distribution Estimation Based on Maximum Entropy Principle Irwin King and Zhong Jin Nov
Reverse Engineering Niloy J. Mitra.
INFORMATIK Differential Coordinates for Interactive Mesh Editing Yaron Lipman Olga Sorkine Daniel Cohen-Or David Levin Tel-Aviv University Christian Rössl.
Mean Shift A Robust Approach to Feature Space Analysis Kalyan Sunkavalli 04/29/2008 ES251R.
CENG 789 – Digital Geometry Processing 06- Rigid-Body Alignment Asst. Prof. Yusuf Sahillioğlu Computer Eng. Dept,, Turkey.
Example Based 3D Shape Completion Mark Pauly 1,2, Niloy J. Mitra 1, Joachim Giesen 2, Markus Gross 2, Leonidas J. Guibas 1 1 Stanford University 2 ETH,
Uncertainty and Variability in Point Cloud Surface Data Mark Pauly 1,2, Niloy J. Mitra 1, Leonidas J. Guibas 1 1 Stanford University 2 ETH, Zurich.
Iterative closest point algorithms
Pauly, Keiser, Kobbelt, Gross: Shape Modeling with Point-Sampled GeometrySIGGRAPH 2003 Shape Modeling with Point-Sampled Geometry Mark Pauly Richard Keiser.
© University of Minnesota Data Mining for the Discovery of Ocean Climate Indices 1 CSci 8980: Data Mining (Fall 2002) Vipin Kumar Army High Performance.
Segmentation Divide the image into segments. Each segment:
Correspondence & Symmetry
1 Numerical geometry of non-rigid shapes Spectral Methods Tutorial. Spectral Methods Tutorial 6 © Maks Ovsjanikov tosca.cs.technion.ac.il/book Numerical.
Point Based Animation of Elastic, Plastic and Melting Objects Matthias Müller Richard Keiser Markus Gross Mark Pauly Andrew Nealen Marc Alexa ETH Zürich.
Niloy J. Mitra1, Natasha Gelfand1, Helmut Pottmann2, Leonidas J
Recovering Articulated Object Models from 3D Range Data Dragomir Anguelov Daphne Koller Hoi-Cheung Pang Praveen Srinivasan Sebastian Thrun Computer Science.
Spectral Processing of Point-sampled Geometry
Matching Compare region of image to region of image. –We talked about this for stereo. –Important for motion. Epipolar constraint unknown. But motion small.
Automatic Image Alignment (feature-based) : Computational Photography Alexei Efros, CMU, Fall 2006 with a lot of slides stolen from Steve Seitz and.
The Planar-Reflective Symmetry Transform Princeton University.
Relevance Feedback Content-Based Image Retrieval Using Query Distribution Estimation Based on Maximum Entropy Principle Irwin King and Zhong Jin The Chinese.
Meshless Animation of Fracturing Solids Mark Pauly Leonidas J. Guibas Richard Keiser Markus Gross Bart Adams Philip Dutré.
כמה מהתעשייה? מבנה הקורס השתנה Computer vision.
Image Segmentation Rob Atlas Nick Bridle Evan Radkoff.
Mean-shift and its application for object tracking
Mean Shift Theory and Applications Reporter: Zhongping Ji.
Shape Matching for Model Alignment 3D Scan Matching and Registration, Part I ICCV 2005 Short Course Michael Kazhdan Johns Hopkins University.
FlowString: Partial Streamline Matching using Shape Invariant Similarity Measure for Exploratory Flow Visualization Jun Tao, Chaoli Wang, Ching-Kuang Shene.
Local invariant features Cordelia Schmid INRIA, Grenoble.
7.1. Mean Shift Segmentation Idea of mean shift:
City University of Hong Kong 18 th Intl. Conf. Pattern Recognition Self-Validated and Spatially Coherent Clustering with NS-MRF and Graph Cuts Wei Feng.
ALIGNMENT OF 3D ARTICULATE SHAPES. Articulated registration Input: Two or more 3d point clouds (possibly with connectivity information) of an articulated.
Detecting Curved Symmetric Parts using a Deformable Disc Model Tom Sie Ho Lee, University of Toronto Sanja Fidler, TTI Chicago Sven Dickinson, University.
CSE 185 Introduction to Computer Vision Pattern Recognition 2.
CSCE 643 Computer Vision: Structure from Motion
Enforcing Constraints for Human Body Tracking David Demirdjian Artificial Intelligence Laboratory, MIT.
INFORMATIK Laplacian Surface Editing Olga Sorkine Daniel Cohen-Or Yaron Lipman Tel Aviv University Marc Alexa TU Darmstadt Christian Rössl Hans-Peter Seidel.
EECS 274 Computer Vision Segmentation by Clustering II.
Andrew Nealen / Olga Sorkine / Mark Alexa / Daniel Cohen-Or SoHyeon Jeong 2007/03/02.
CS654: Digital Image Analysis Lecture 30: Clustering based Segmentation Slides are adapted from:
CS654: Digital Image Analysis
Mesh Coarsening zhenyu shu Mesh Coarsening Large meshes are commonly used in numerous application area Modern range scanning devices are used.
Temporally Coherent Completion of Dynamic Shapes AUTHORS:HAO LI,LINJIE LUO,DANIEL VLASIC PIETER PEERS,JOVAN POPOVIC,MARK PAULY,SZYMON RUSINKIEWICZ Presenter:Zoomin(Zhuming)
Gaussian Mixture Models and Expectation-Maximization Algorithm.
Image Segmentation Shengnan Wang
Mean Shift ; Theory and Applications Presented by: Reza Hemati دی 89 December گروه بینایی ماشین و پردازش تصویر Machine Vision and Image Processing.
A D V A N C E D C O M P U T E R G R A P H I C S CMSC 635 January 15, 2013 Quadric Error Metrics 1/20 Geometric Morphometrics Feb 27, 2013 Geometric Morphologyd.
CAD Mesh Model Segmentation by Clustering
Robust Estimation Course web page: vision.cis.udel.edu/~cv April 23, 2003  Lecture 25.
CENG 789 – Digital Geometry Processing 07- Rigid-Body Alignment Asst. Prof. Yusuf Sahillioğlu Computer Eng. Dept,, Turkey.
Point Based Animation of Elastic, Plastic and Melting Objects Mark Pauly Andrew Nealen Marc Alexa ETH Zürich TU Darmstadt Stanford Matthias Müller Richard.
Color Image Segmentation Mentor : Dr. Rajeev Srivastava Students: Achit Kumar Ojha Aseem Kumar Akshay Tyagi.
MASKS © 2004 Invitation to 3D vision Lecture 3 Image Primitives andCorrespondence.
Shape2Pose: Human Centric Shape Analysis CMPT888 Vladimir G. Kim Siddhartha Chaudhuri Leonidas Guibas Thomas Funkhouser Stanford University Princeton University.
Invariant Local Features Image content is transformed into local feature coordinates that are invariant to translation, rotation, scale, and other imaging.
Physically-Based Motion Synthesis in Computer Graphics
Semi-Supervised Clustering
CENG 789 – Digital Geometry Processing 08- Rigid-Body Alignment
Qualitative Curve Descriptions
Nonparametric Semantic Segmentation
You can check broken videos in this slide here :
Spectral Methods Tutorial 6 1 © Maks Ovsjanikov
Image Stitching Slides from Rick Szeliski, Steve Seitz, Derek Hoiem, Ira Kemelmacher, Ali Farhadi.
CENG 789 – Digital Geometry Processing 09- Rigid-Body Alignment
Presentation transcript:

Niloy J. Mitra Leonidas J. Guibas Mark Pauly TU Vienna Stanford University ETH Zurich SIGGRAPH 2007

 Invariance under a class of transformations 2

 Goal: Symmetrize 3D geometry  Approach: Minimally deform the model in the spatial domain by optimizing the distribution in transformation space 3

 Given an explicit point ‐ pairing, a closed form solution for symmetrizing the point set  A symmetrization algorithm that uses transform domain reasoning to guide shape deformation in object domain  Applications: ◦ Extend the types of detected symmetries ◦ Symmetric remeshing ◦ Automatic correspondence for articulated bodies 4

 Mitra, Guibas, Pauly: Partial and Approximate Symmetry Detection for 3d Geometry. ACM Trans. Graph. 25, 3,

 Initial pairs are sampled randomly  Pruning based on curvature and normal 6

 Use mean-shift algorithm ◦ Non-Parametric Density Estimation 7 The blue data points were traversed by the windows towards the mode

 Goal : Extracting the connected components of the model from cluster  Starting with a random point of cluster ◦ Corresponds to a pair (p i, p j ) of points on the model surface  Look at the one-ring neighbors p i and apply T  Check distances of the transformed points to the surface around p j 8

Transformation space d 9

10

 Pairs of sample points define reflective symmetry transform 11

 Density plot → accumulation of symmetry evidence 12

 Density cluster → reflective symmetry 13

14

A set of potential corresponding point pairs extracted 15

16

Cluster contraction Local symmetrization Cluster contraction in transform space Constrained deformation in object space 17

 Object space point pairs → points in transform space  Cluster in transform space corresponds to approximate symmetry  Cluster contraction in transform space corresponds to constrained in deformation in object space that enhances object symmetry 18

Cluster merging → global symmetrization 19

Cluster merging/contraction → Global symmetrization 20

 Local Symmetrization ◦ Cluster contraction How to deform in the spatial domain ? Where to move in transform space ?  Global Symmetrization ◦ Cluster merging 21

 Goal: Minimally displace two points to make them symmetric with respect to a given transformation [Zabrodsky et al. 1997] 22

 Goal: Find optimal transformation and minimal displacements for a set of point ‐ pairs 23

 Reflection ◦ Minimize energy ◦ Reduced to eigenvalue problem  Rigid Transform ◦ Minimize energy ◦ SVD problem 24

 Initial random sampling does not respect symmetries.  The correspondences estimated during the symmetry detection stage are potentially inaccurate and incomplete 25

 Every sample p shifted in the direction of displacement d p (white circle)  Project them onto the surface (colored square)  The procedure is iterated until the variance of the cluster is no longer reduced. 26

 Local Symmetrization ◦ Cluster contraction Where to move in transform space ? How to deform in the spatial domain ?  Optimal transformation  Global Symmetrization ◦ Cluster merging 27

 Using existing shape deformation method ◦ Symmetrizing displacements  positional constraints ◦ 2D : As-rigid-as-possible shape manipulation method[Igarashi et al.2005] ◦ 3D : Non-linear PriMo deformation model [Botsch et al. 2006] 28 As-Rigid-As-Possible Shape Manipulation [Igarashi 2005] PriMo: Coupled Prisms for Intuitive Surface Modeling [Botsch 2006]

 Find sample pairs  Optimize sample positions on surface  Compute the optimal transformation  Update p i :  p are used as deformation constraints  Re-compute the optimal transformation  Find new sample pairs every 5 time step 29

 Sort clusters by height  Select the most pronounced cluster for symmetrization  Apply the symmetrizing deformation  Repeat the process with next biggest cluster  Finally, Merge clusters based on distance greedily 30

 User controls the deformation by modifying the stiffness of the shape’s material  Soft materials allow for better symmetrization  Stiffer materials more strongly resist the symmetrizing deformation  System allow spatially varying stiffness  User controls the symmetrization by interactively selecting clusters for contraction or merging 31

32

33

34 Symmetry Based Remeshing [Podolak al SGP 2007]

35

36

37

 Some case, method is fails to process the entire model ◦ The front feet of the bunny and the right foot of the male character  Small-scale features are sometimes ignored  Insufficient local matching  The deformation model does not respect the semantics of the shape. 38

 Symmetrization algorithm ◦ Robust and efficient, requires minimal user intervention ◦ Handle both local and global symmetries  Future Work ◦ Symmetry respecting geometry processing ◦ Hierarchical shape semantics ◦ Perception, art, design ◦ Other data, e.g. motion data, derived spaces 39