Texture Synthesis from Multiple Sources Li-Yi Wei Stanford University (was) NVIDIA Corporation (now)

Slides:



Advertisements
Similar presentations
Eurographics 2012, Cagliari, Italy Template for EG2012 presentations CRS4 EG2012 Team.
Advertisements

Inverse Texture Synthesis Li-Yi Wei 1 Jianwei Han 2 Kun Zhou 1,2 Hujun Bao 2 Baining Guo 1 Harry Shum 1 1 Microsoft 2 Zhejiang University.
Lapped textures Emil Praun Adam Finkelstein Hugues Hoppe
Pattern-based Texture Metamorphosis Z. Liu, C. Liu, and H. Shum Microsoft Research Asia Y. Yu UIUC.
Tile-Based Texture Mapping on Graphics Hardware Li-Yi Wei NVIDIA.
Andrew Nealen and Marc Alexa, Discrete Geometric Modeling Group, TU Darmstadt, 2004 Fast and High Quality Overlap Repair for Patch-Based Texture Synthesis.
Super-Resolution Texturing for Online Virtual Globes
Texture. Limitation of pixel based processing Edge detection with different threshold.
Data-driven methods: Texture (Sz 10.5) Cs129 Computational Photography James Hays, Brown, Spring 2011 Many slides from Alexei Efros.
Texture Synthesis on [Arbitrary Manifold] Surfaces Presented by: Sam Z. Glassenberg* * Several slides borrowed from Wei/Levoy presentation.
Procedural Content Tiling
Texture. Edge detectors find differences in overall intensity. Average intensity is only simplest difference. many slides from David Jacobs.
Edit this text to create a title slide And put your sub-heading here.
Texture Synthesis Tiantian Liu. Definition Texture – Texture refers to the properties held and sensations caused by the external surface of objects received.
Lapped Textures Emil Praun Adam Finkelstein Hugues Hoppe Emil Praun Adam Finkelstein Hugues Hoppe Princeton University Microsoft Research Princeton University.
Image Quilting for Texture Synthesis & Transfer Alexei Efros (UC Berkeley) Bill Freeman (MERL) +=
Texture Synthesis by Non-parametric Sampling / Image Quilting for Texture Synthesis & Transfer by Efros and Leung / Efros and Freeman ICCV ’99 / SIGGRAPH.
Overview of Texture Synthesis Ganesh Ramanarayanan Cornell Graphics Seminar.
Order-Independent Texture Synthesis Li-Yi Wei Marc Levoy Gcafe 1/30/2003.
Fast Texture Synthesis using Tree-structured Vector Quantization Li-Yi Wei Marc Levoy Computer Graphics Group Stanford University.
Direct Methods for Visual Scene Reconstruction Paper by Richard Szeliski & Sing Bing Kang Presented by Kristin Branson November 7, 2002.
A Crash Course on Programmable Graphics Hardware Li-Yi Wei 2005 at Tsinghua University, Beijing.
A Crash Course on Texturing Li-Yi Wei 2005 at Tsinghua University, Beijing.
Texture Splicing Yiming Liu, Jiaping Wang, Su Xue, Xin Tong, Sing Bing Kang, Baining Guo.
Image-Based Rendering Produce a new image from real images. Combining images Interpolation More exotic methods.
Texture Synthesis on Surfaces Paper by Greg Turk Presentation by Jon Super.
Announcements For future problems sets: matlab code by 11am, due date (same as deadline to hand in hardcopy). Today’s reading: Chapter 9, except.
Texture Reading: Chapter 9 (skip 9.4) Key issue: How do we represent texture? Topics: –Texture segmentation –Texture-based matching –Texture synthesis.
Andrew Nealen and Marc Alexa, Discrete Geometric Modeling Group, TU Darmstadt, 2003 Hybrid Texture Synthesis Andrew Nealen Marc Alexa Discrete Geometric.
Stereological Techniques for Solid Textures Rob Jagnow MIT Julie Dorsey Yale University Holly Rushmeier Yale University.
Texture Synthesis over Arbitrary Manifold Surfaces Li-Yi Wei Marc Levoy Computer Graphics Group Stanford University.
Texture perception Lavanya Sharan February 23rd, 2011.
HP - PURDUE CONFIDENTIAL Slide No. Fundamentals 1.
SIGGRAPH 2003 Jingdan Zhang, Kun Zhou, Luiz Velho, Baining Guo, Heung-Yeung Shum.
Efficient Editing of Aged Object Textures By: Olivier Clément Jocelyn Benoit Eric Paquette Multimedia Lab.
CAP5415: Computer Vision Lecture 4: Image Pyramids, Image Statistics, Denoising Fall 2006.
Video Textures Arno Schödl Richard Szeliski David Salesin Irfan Essa Microsoft Research, Georgia Tech.
Edit this text to create a Heading  This subtitle is 20 points  Bullets are blue  They have 110% line spacing, 2 points before & after  Longer bullets.
Continuous Model Synthesis Paul Merrell and Dinesh Manocha In SIGGRAPH Asia 2008 발표 : 이성호.
Texture Optimization for Example-based Synthesis Vivek Kwatra Irfan Essa Aaron Bobick Nipun Kwatra.
Efficient Rendering of Local Subsurface Scattering Tom Mertens 1, Jan Kautz 2, Philippe Bekaert 1, Frank Van Reeth 1, Hans-Peter Seidel
Image Quilting for Texture Synthesis and Transfer Alexei A. Efros (UC Berkeley) William T. Freeman (MERL) Siggraph01 ’
Title Slide (Arial bold, 36 pt., centered ) Name of Presenter (Arial bold, 24 pt., centered)
Lapped Solid Textures: Filling a Model with Anisotropic Textures Kenshi Takayama 1 Makoto Okabe 1 Takashi Ijiri 1 Takeo Igarashi 1,2 1 The University of.
Stereological Techniques for Solid Textures Rob Jagnow MIT Julie Dorsey Yale University Holly Rushmeier Yale University.
Synthesis of Compact Textures for real-time Terrain Rendering Nader Salman 22 juin 2007 Encadrant : Sylvain Lefebvre.
Edit this text to create a Heading  This subtitle is 20 points  Bullets are blue  They have 110% line spacing, 2 points before & after  Longer bullets.
TextureAmendment Reducing Texture Distortion in Constrained Parameterizations Yu-Wing TaiNational University of Singapore Michael S. BrownNational University.
Towards Real-Time Texture Synthesis With the Jump Map Steve Zelinka Michael Garland University of Illinois at Urbana-Champaign Thirteenth Eurographics.
2D Texture Synthesis Instructor: Yizhou Yu. Texture synthesis Goal: increase texture resolution yet keep local texture variation.
BEST PRACTICES FOR USAGE Text Use minimal text; presentations should be visual in nature The presentation uses NIST's official fonts; please try to avoiding.
Edit this text to create a Heading  This subtitle is 20 points  Bullets are blue  They have 110% line spacing, 2 points before & after  Longer bullets.
Two Patch-based Algorithms for By-example Texture Synthesis Bruno Galerne MAP5, Université Paris Descartes 1 Master 2 Traitement.
Geometry Synthesis Ares Lagae Olivier Dumont Philip Dutré Department of Computer Science Katholieke Universiteit Leuven 10 August, 2004.
Data-driven Architectural texture mapping Texture mapping Un-textured 3D sceneTextured output Textured Architectures 由于建筑物的3D model和 textures均属于structured.
Multi-Class Blue Noise Sampling Li-Yi Wei Microsoft Research.
Video Textures Arno Schödl Richard Szeliski David Salesin Irfan Essa Microsoft Research, Georgia Tech.
Tracking Hands with Distance Transforms Dave Bargeron Noah Snavely.
Texture Analysis and Synthesis. Texture Texture: pattern that “looks the same” at all locationsTexture: pattern that “looks the same” at all locations.
Announcements Project 4 out today help session at the end of class.
Powerpoint lesson 3- working with text
Detail Preserving Shape Deformation in Image Editing
Context-based vision system for place and object recognition
Texture.
Image Quilting for Texture Synthesis & Transfer
Announcements Guest lecture next Tuesday
Edit this text to create a title slide
Title of Presentation Verdana 40pt Bold
Sample text or Subtitle
Sample text or Subtitle
Presentation transcript:

Texture Synthesis from Multiple Sources Li-Yi Wei Stanford University (was) NVIDIA Corporation (now)

Texture Synthesis (Single-Source) Input Output Synthesis Tong et. al Soler et. al Hertzmann et. al Efros & Freeman 2001 Yin et. al Turk 2001 Wei & Levoy 2001 Wei & Levoy 2000 Portilla & Simoncelli 1999 Efros & Leung 1999 Heeger & Bergen 1995 De Bonet 1997

Limitations of Single-Source Synthesis Non-uniform, varying patterns –Junction of 2 textures –Varying scale, orientation, color, shape –Creating new textures Textures of different dimensions –Solid textures from 2D views

Multiple-Source Texture Synthesis Solid textures from multiple 2D views View 3 View 2 View 1 Texture mixtures from multiple sources += ? Source 1Source 2

Previous Work Solid texture from 2D views –Heeger & Bergen 1995 –Ghazanfarpour & Dischler 1999 Images from [Heeger&Bergen 1995] 2D Source3D result

Previous Work Texture varying, morphing, and mixture –Portilla & Simoncelli 1999 –Bar-Joseph, El-Yaniv, Lichinski, Werman 2001 –Z. Liu, C. Liu, Shum, Yu 2003 –PVT paper in SIGGRAPH 2003 Image from [Portilla&Simoncelli 1999] Source 1Source 2morphing

Algorithm Source Synthesis Result N(p) search N(p i ) Single-source algorithm [ICCV 99, SIGGRAPH 2000] copy p

Algorithm Source 1 Source 2 Source 3 Synthesis Result search N 1 (p 1 ) N 2 (p 2 ) N 3 (p 3 ) E(p, {p i }) = Σ w i × ( |p-p i | 2 + |N i (p)-N i (p i )| 2 ) i= 1,2,3 L2 distance Fix p, search {p i } Fix {p i }, set p = Σ w i ×p i User weights Average

Algorithm Source 1 Source 2 Source 3 Synthesis Result search E(p, {p i }) = Σ w i × ( |p-p i | 2 + |N i (p)-N i (p i )| 2 ) i= 1,2,3 L2 distance Fix p, search {p i } Fix {p i }, set p = Σ w i ×p i User weights Average

Texture Mixture Example += Uniform Transition Use weights w i to control the result Source 1Source 2

Texture Mixture Results Source1Source2Mixture-uniformMixture-transition

Texture Mixture Results Source1Source2Mixture-uniformMixture-transition

Solid Texture Synthesis View 3 View 2 View 1 Use {N i } with different orientations N2N2 N3N3 N1N1

Specifying Views Input ViewsResult × Inconsistent О О

Solid Texture Results

Comparison : Surface/Volume Synthesis Surface [SIGGRAPH 2001] Volume [this work]

Conclusion Single-source synthesis –mostly mature and well done –limited capability Multiple-source synthesis –hard (patch-copying won’t work!) –more useful –requires more work!

Future Work Combine the texton mask idea in SIGGRAPH 2003 PVT paper with this algorithm?

END

Algorithm 1.Fix p, search {p i } to minimize |N i (p)-N i (p i )| 2 2.Fix {p i }, set p = Σ w i ×p i Goal : Minimize E(p, {p i }) = Σ w i × ( |p-p i | 2 + |N i (p)-N i (p i )| 2 ) Step 1Step 2 How : iterate

Comparison ReusableDistortionQualityEfficiencyTunable Surfacebetter Volumebetter

S2003 Verdana, Bold, 37 points This subtitle is 31 points Bullets are orange; text is 26 points They have 110% line spacing, 6 points before/after Longer bullets in the form of a paragraph are harder to read if there is insufficient line spacing. This is the maximum recommended number of lines per slide (seven). –Sub-bullets look like this. This subtitle is 31 points Bullets are orange; text is 26 points They have 110% line spacing, 6 points before/after Longer bullets in the form of a paragraph are harder to read if there is insufficient line spacing. This is the maximum recommended number of lines per slide (seven). –Sub-bullets look like this.