Image Stitching. Computer Vision Jia-Bin Huang, Virginia Tech. Add example. Many slides from S. Seitz and D. Hoiem

Size: px
Start display at page:

Download "Image Stitching. Computer Vision Jia-Bin Huang, Virginia Tech. Add example. Many slides from S. Seitz and D. Hoiem"

Transcription

1 Iage Stitcing Add exaple Coputer Vision Jia-Bin Huang, Virginia Tec Many slides fro S. Seitz and D. Hoie

2 Adinistrative stuffs HW 3 is out due :59 PM Oct 7 Please start early. Deadlines are fir. No eails requesting extensions Getting elp? *Five* free late days witout penalty Piazza Office ours No free late dates for final projects

3 Review: Caera Projection Matrix Z Y X t r r r t r r r t r r r v f u s f v u w z y x X t x K R O w i w k w j w t R 3

4 Review: Caera Calibration Metod : Use an object (calibration grid) wit known geoetry Correspond iage points to 3d points Get least squares solution (or non-linear solution) Known 2d iage coordinates Known 3d locations wu wv w X Y Z Unknown Caera Paraeters 4

5 Z Y X s sv su Known 3d locations Known 2d iage coords Unknown Caera Paraeters u uz uy ux Z Y X v vz vy vx Z Y X Hoogeneous linear syste. Solve for s entries using linear least squares v Z v Y v X v Z Y X u Z u Y u X u Z Y X v Z v v Y X v Z Y X u Z u u Y X u Z Y X n n n n n n n n n n n n n n n n n n n n [U, S, V] = svd(a); M = V(:,end); M = resape(m,[],3)';

6 Review: Calibration by vanising points VP (2D) VP (3D) Ortogonality constraints X i X j = X i = R K p i K = p i = KRX i p i K R (R )(K )p j = Constraints for p, p 2, p 3 K = f u f v f u f f v f Unknown caera paraeters f, u, v p K (K )p 2 = p K (K )p 3 = p 2 K (K )p 3 = x u x 2 u + y v y 2 v + f 2 = Eqn () x u x 3 u + y v y 3 v + f 2 = Eqn (2) x 2 u x 3 u + y 2 v y 3 v + f 2 = Eqn (3) Eqn () Eqn (2) x u x 2 x 3 + y v y 2 y 3 = Eqn (2) Eqn (3) x 3 u x x 2 + y 3 v y y 2 = Solve for u, v f = x u x 2 u y v y 2 v

7 Review: Calibration by vanising points Rotation atrix R = r r 2 r 3 Unknown caera paraeters R p i = KRX i Set directions of vanising points X =,, X 2 =,, X 3 =,, p = Kr p 2 = Kr 2 p 3 = Kr 3 r = K p r 2 = K p 2 r 3 = K p 3 Special properties of R inv(r)=r T Eac row and colun of R as unit lengt

8 Measuring eigt v z r Slide by Steve Seitz vanising line (orizon) v x v t H t R H v y b b t b r b v v Z Z r t iage cross ratio H R 8

9 Tis class: Iage Stitcing Cobine two or ore overlapping iages to ake one larger iage Add exaple Slide credit: Vaibav Vais

10 Concepts introduced/reviewed in today s lecture Caera odel Hoograpies Solving oogeneous systes of linear equations Keypoint-based alignent RANSAC Blending How te ipone stitcer works

11 Illustration Caera Center

12 Proble set-up x = K [R t] X x' = K' [R' t'] X t=t'= x'. X x f f' x'=hx were H = K' R' R - K - Typically only R and f will cange (4 paraeters), but, in general, H as 8 paraeters

13 Hoograpy Definition General ateatics: oograpy = projective linear transforation Vision (ost coon usage): oograpy = linear transforation between iage planes two Exaples Project 3D surface into frontal view Relate two views tat differ only by rotation

14 Hoograpy exaple: Iage rectification p p To unwarp (rectify) an iage solve for oograpy H given p and p : wp =Hp

15 Hoograpy exaple: Planar apping Freedo HP Coercial

16 Iage Stitcing Algorit Overview. Detect keypoints (e.g., SIFT) 2. Matc keypoints (e.g., st /2 nd NN < tres) 3. Estiate oograpy wit four atced keypoints (using RANSAC) 4. Cobine iages

17 Coputing oograpy Assue we ave four atced points: How do we copute oograpy H? Direct Linear Transforation (DLT) x Hx ' v vv uv v u u vu uu v u ' ' ' ' ' ' w w v w u x H

18 Coputing oograpy Direct Linear Transfor Apply SVD: UDV T = A = V sallest (colun of V corr. to sallest singular value) H A n n n n n n n v v v v u v u v v v v u v u u u v u u v u Matlab [U, S, V] = svd(a); = V(:, end); Explanations of SVD and solving oogeneous linear systes

19 Coputing oograpy Assue we ave four atced points: How do we copute oograpy H? Noralized DLT. Noralize coordinates for eac iage a) Translate for zero ean b) Scale so tat average distance to origin is ~sqrt(2) ~ x Tx ~ x T x Tis akes proble better beaved nuerically (see HZ p. 7-8) 2. Copute H ~ using DLT in noralized coordinates 3. Unnoralize: H T ~ HT x i Hx i

20 HZ Tutorial 99 Coputing oograpy Assue we ave atced points wit outliers: How do we copute oograpy H? Autoatic Hoograpy Estiation wit RANSAC. Coose nuber of saples N

21 HZ Tutorial 99 Coputing oograpy Assue we ave atced points wit outliers: How do we copute oograpy H? Autoatic Hoograpy Estiation wit RANSAC. Coose nuber of saples N 2. Coose 4 rando potential atces 3. Copute H using noralized DLT 4. Project points fro x to x for eac potentially atcing pair: xi Hx i 5. Count points wit projected distance < t E.g., t = 3 pixels 6. Repeat steps 2-5 N ties Coose H wit ost inliers

22 Autoatic Iage Stitcing. Copute interest points on eac iage 2. Find candidate atces 3. Estiate oograpy H using atced points and RANSAC wit noralized DLT 4. Project eac iage onto te sae surface and blend Matlab: aketfor, itransfor

23 RANSAC for Hoograpy Initial Matced Points

24 RANSAC for Hoograpy Final Matced Points

25 RANSAC for Hoograpy

26 Coosing a Projection Surface Many to coose: planar, cylindrical, sperical, cubic, etc.

27 Planar Mapping x x f f ) For red iage: pixels are already on te planar surface 2) For green iage: ap to first iage plane

28 Planar Projection Planar Potos by Russ Hewett

29 Planar Projection Planar

30 Cylindrical Mapping x x f f ) For red iage: copute, teta on cylindrical surface fro (u, v) 2) For green iage: ap to first iage plane, tan ap to cylindrical surface

31 Cylindrical Projection Cylindrical

32 Cylindrical Projection Cylindrical

33 Planar Cylindrical

34 Recognizing Panoraas Soe of following aterial fro Brown and Lowe 23 talk Brown and Lowe 23, 27

35 Recognizing Panoraas Input: N iages. Extract SIFT points, descriptors fro all iages 2. Find K-nearest neigbors for eac point (K=4) 3. For eac iage a) Select M candidate atcing iages by counting atced keypoints (=6) b) Solve oograpy H ij for eac atced iage

36 Recognizing Panoraas Input: N iages. Extract SIFT points, descriptors fro all iages 2. Find K-nearest neigbors for eac point (K=4) 3. For eac iage a) Select M candidate atcing iages by counting atced keypoints (=6) b) Solve oograpy H ij for eac atced iage c) Decide if atc is valid (n i > n f ) # inliers # keypoints in overlapping area

37 Recognizing Panoraas (cont.) (now we ave atced pairs of iages) 4. Find connected coponents

38 Finding te panoraas

39 Finding te panoraas

40 Recognizing Panoraas (cont.) (now we ave atced pairs of iages) 4. Find connected coponents 5. For eac connected coponent a) Perfor bundle adjustent to solve for rotation (θ, θ 2, θ 3 ) and focal lengt f of all caeras b) Project to a surface (plane, cylinder, or spere) c) Render wit ultiband blending

41 Bundle adjustent for stitcing Non-linear iniization of re-projection error xˆ Hx were H = K R R - K - error N i M j k dist( x, xˆ ) Solve non-linear least squares (Levenberg- Marquardt algorit) See paper for details

42 Bundle Adjustent New iages initialised wit rotation, focal lengt of best atcing iage

43 Bundle Adjustent New iages initialised wit rotation, focal lengt of best atcing iage

44 Details to ake it look good Coosing seas Blending

45 Coosing seas Easy etod Assign eac pixel to iage wit nearest center i i2 x x Iage 2 Iage

46 Coosing seas Easy etod Assign eac pixel to iage wit nearest center Create a ask: ask(y, x) = iff pixel sould coe fro i Soot boundaries (called featering ): ask_s = ifilter(ask, gausfil); Coposite iblend = i_c.*ask + i2_c.*(-ask); i i2 x x Iage 2 Iage

47 Coosing seas Better etod: dynaic progra to find sea along well-atced regions Illustration: ttp://en.wikipedia.org/wiki/file:rocester_ny.jpg

48 Gain copensation Siple gain adjustent Copute average RGB intensity of eac iage in overlapping region Noralize intensities by ratio of averages

49 Multi-band Blending Burt & Adelson 983 Blend frequency bands over range l

50 Multiband Blending wit Laplacian Pyraid At low frequencies, blend slowly At ig frequencies, blend quickly Left pyraid blend Rigt pyraid

51 Multiband blending Laplacian pyraids. Copute Laplacian pyraid of iages and ask 2. Create blended iage at eac level of pyraid 3. Reconstruct coplete iage

52 Blending coparison (IJCV 27)

53 Blending Coparison

54 Furter reading DLT algorit: HZ p. 9 (alg 4.2), p. 585 Noralization: HZ p. 7-9 (alg 4.2) RANSAC: HZ Sec 4.7, p. 23, alg 4.6 Rick Szeliski s alignent/stitcing tutorial Recognising Panoraas: Brown and Lowe, IJCV 27 (also bundle adjustent)

55 How does ipone panoraic stitcing work? Capture iages at 3 fps Stitc te central /8 of a selection of iages Select wic iages to stitc using te acceleroeter and frae-tofrae atcing Faster and avoids radial distortion tat often occurs towards corners of iages Alignent Initially, perfor cross-correlation of sall patces aided by acceleroeter to find good regions for atcing Register by atcing points (KLT tracking or RANSAC wit FAST (siilar to SIFT) points) or correlational atcing Blending Linear (or siilar) blending, using a face detector to avoid blurring face regions and coose good face sots (not blinking, etc) ttp://

56 Tings to reeber Hoograpy relates rotating caeras Recover oograpy using RANSAC and noralized DLT Bundle adjustent iniizes reprojection error for set of related iages Details to ake it look nice (e.g., blending)

57 See you on Trusday Next class: Epipolar Geoetry and Stereo Vision

Colorado School of Mines. Computer Vision. Professor William Hoff Dept of Electrical Engineering &Computer Science.

Colorado School of Mines. Computer Vision. Professor William Hoff Dept of Electrical Engineering &Computer Science. Professor Willia Hoff Dept of Electrical Engineering &Coputer Science http://inside.ines.edu/~whoff/ 1 Caera Calibration 2 Caera Calibration Needed for ost achine vision and photograetry tasks (object

More information

Image Stitching. Slides from Rick Szeliski, Steve Seitz, Derek Hoiem, Ira Kemelmacher, Ali Farhadi

Image Stitching. Slides from Rick Szeliski, Steve Seitz, Derek Hoiem, Ira Kemelmacher, Ali Farhadi Image Stitching Slides from Rick Szeliski, Steve Seitz, Derek Hoiem, Ira Kemelmacher, Ali Farhadi Combine two or more overlapping images to make one larger image Add example Slide credit: Vaibhav Vaish

More information

(Geometric) Camera Calibration

(Geometric) Camera Calibration (Geoetric) Caera Calibration CS635 Spring 217 Daniel G. Aliaga Departent of Coputer Science Purdue University Caera Calibration Caeras and CCDs Aberrations Perspective Projection Calibration Caeras First

More information

Feature Based Registration for Panoramic Image Generation

Feature Based Registration for Panoramic Image Generation IJCSI International Journal of Coputer Science Issues, Vol. 10, Issue 6, No, Noveber 013 www.ijcsi.org 13 Feature Based Registration for Panoraic Iage Generation Kawther Abbas Sallal 1, Abdul-Mone Saleh

More information

Epipolar Geometry and Stereo Vision

Epipolar Geometry and Stereo Vision Epipolar Geometry and Stereo Vision Computer Vision Jia-Bin Huang, Virginia Tech Many slides from S. Seitz and D. Hoiem Last class: Image Stitching Two images with rotation/zoom but no translation. X x

More information

Affine and Projective Transformations

Affine and Projective Transformations CS 674: Intro to Computer Vision Affine and Projective Transformations Prof. Adriana Kovaska Universit of Pittsburg October 3, 26 Alignment problem We previousl discussed ow to matc features across images,

More information

Local features and image matching May 8 th, 2018

Local features and image matching May 8 th, 2018 Local features and image matcing May 8 t, 2018 Yong Jae Lee UC Davis Last time RANSAC for robust fitting Lines, translation Image mosaics Fitting a 2D transformation Homograpy 2 Today Mosaics recap: How

More information

A simplified approach to merging partial plane images

A simplified approach to merging partial plane images A siplified approach to erging partial plane iages Mária Kruláková 1 This paper introduces a ethod of iage recognition based on the gradual generating and analysis of data structure consisting of the 2D

More information

Homographies and RANSAC

Homographies and RANSAC Homographies and RANSAC Computer vision 6.869 Bill Freeman and Antonio Torralba March 30, 2011 Homographies and RANSAC Homographies RANSAC Building panoramas Phototourism 2 Depth-based ambiguity of position

More information

MAPI Computer Vision

MAPI Computer Vision MAPI Computer Vision Multiple View Geometry In tis module we intend to present several tecniques in te domain of te 3D vision Manuel Joao University of Mino Dep Industrial Electronics - Applications -

More information

Feature Matching and RANSAC

Feature Matching and RANSAC Feature Matching and RANSAC Recognising Panoramas. [M. Brown and D. Lowe,ICCV 2003] [Brown, Szeliski, Winder, CVPR 2005] with a lot of slides stolen from Steve Seitz, Rick Szeliski, A. Efros Introduction

More information

Single-view 3D Reconstruction

Single-view 3D Reconstruction Single-view 3D Reconstruction 10/12/17 Computational Photography Derek Hoiem, University of Illinois Some slides from Alyosha Efros, Steve Seitz Notes about Project 4 (Image-based Lighting) You can work

More information

Image-based Modeling and Rendering: 8. Image Transformation and Panorama

Image-based Modeling and Rendering: 8. Image Transformation and Panorama Image-based Modeling and Rendering: 8. Image Transformation and Panorama I-Chen Lin, Assistant Professor Dept. of CS, National Chiao Tung Univ, Taiwan Outline Image transformation How to represent the

More information

Object Recognition with Invariant Features

Object Recognition with Invariant Features Object Recognition with Invariant Features Definition: Identify objects or scenes and determine their pose and model parameters Applications Industrial automation and inspection Mobile robots, toys, user

More information

Clustering. Cluster Analysis of Microarray Data. Microarray Data for Clustering. Data for Clustering

Clustering. Cluster Analysis of Microarray Data. Microarray Data for Clustering. Data for Clustering Clustering Cluster Analysis of Microarray Data 4/3/009 Copyright 009 Dan Nettleton Group obects that are siilar to one another together in a cluster. Separate obects that are dissiilar fro each other into

More information

Undergrad HTAs / TAs. Help me make the course better! HTA deadline today (! sorry) TA deadline March 21 st, opens March 15th

Undergrad HTAs / TAs. Help me make the course better! HTA deadline today (! sorry) TA deadline March 21 st, opens March 15th Undergrad HTAs / TAs Help me make the course better! HTA deadline today (! sorry) TA deadline March 2 st, opens March 5th Project 2 Well done. Open ended parts, lots of opportunity for mistakes. Real implementation

More information

Computer Aided Drafting, Design and Manufacturing Volume 26, Number 2, June 2016, Page 13

Computer Aided Drafting, Design and Manufacturing Volume 26, Number 2, June 2016, Page 13 Coputer Aided Drafting, Design and Manufacturing Volue 26, uber 2, June 2016, Page 13 CADDM 3D reconstruction of coplex curved objects fro line drawings Sun Yanling, Dong Lijun Institute of Mechanical

More information

Epipolar Geometry and Stereo Vision

Epipolar Geometry and Stereo Vision Epipolar Geometry and Stereo Vision Computer Vision Shiv Ram Dubey, IIIT Sri City Many slides from S. Seitz and D. Hoiem Last class: Image Stitching Two images with rotation/zoom but no translation. X

More information

CF37 1DL. 1. Introduction. 2. The Pose-clustering Paradigm

CF37 1DL. 1. Introduction. 2. The Pose-clustering Paradigm 3-D Object Recognition and Orientation fro bot Noisy and Occluded 2-D Data D.P.IHing*, P.T.Fairney + & R.J.Wiltsire,* *Dept. of Mateatics and Coputing, +Dept. of Coputer Studies, Te Polytecnic of Wales,

More information

CS6670: Computer Vision

CS6670: Computer Vision CS6670: Computer Vision Noah Snavely Lecture 7: Image Alignment and Panoramas What s inside your fridge? http://www.cs.washington.edu/education/courses/cse590ss/01wi/ Projection matrix intrinsics projection

More information

Resolution. Super-Resolution Imaging. Problem

Resolution. Super-Resolution Imaging. Problem Resolution Super-Resolution Iaging Resolution: Sallest easurable detail in a visual presentation Subhasis Chaudhuri Departent of Electrical Engineering Indian institute of Technology Bobay Powai, Mubai-400

More information

Image stitching. Digital Visual Effects Yung-Yu Chuang. with slides by Richard Szeliski, Steve Seitz, Matthew Brown and Vaclav Hlavac

Image stitching. Digital Visual Effects Yung-Yu Chuang. with slides by Richard Szeliski, Steve Seitz, Matthew Brown and Vaclav Hlavac Image stitching Digital Visual Effects Yung-Yu Chuang with slides by Richard Szeliski, Steve Seitz, Matthew Brown and Vaclav Hlavac Image stitching Stitching = alignment + blending geometrical registration

More information

Evaluation of a multi-frame blind deconvolution algorithm using Cramér-Rao bounds

Evaluation of a multi-frame blind deconvolution algorithm using Cramér-Rao bounds Evaluation of a ulti-frae blind deconvolution algorith using Craér-Rao bounds Charles C. Beckner, Jr. Air Force Research Laboratory, 3550 Aberdeen Ave SE, Kirtland AFB, New Mexico, USA 87117-5776 Charles

More information

COLOR HISTOGRAM AND DISCRETE COSINE TRANSFORM FOR COLOR IMAGE RETRIEVAL

COLOR HISTOGRAM AND DISCRETE COSINE TRANSFORM FOR COLOR IMAGE RETRIEVAL COLOR HISTOGRAM AND DISCRETE COSINE TRANSFORM FOR COLOR IMAGE RETRIEVAL 1 Te-Wei Chiang ( 蔣德威 ), 2 Tienwei Tsai ( 蔡殿偉 ), 3 Jeng-Ping Lin ( 林正平 ) 1 Dept. of Accounting Inforation Systes, Chilee Institute

More information

Automatic Image Alignment

Automatic Image Alignment Automatic Image Alignment with a lot of slides stolen from Steve Seitz and Rick Szeliski Mike Nese CS194: Image Manipulation & Computational Photography Alexei Efros, UC Berkeley, Fall 2018 Live Homography

More information

Image Filter Using with Gaussian Curvature and Total Variation Model

Image Filter Using with Gaussian Curvature and Total Variation Model IJECT Vo l. 7, Is s u e 3, Ju l y - Se p t 016 ISSN : 30-7109 (Online) ISSN : 30-9543 (Print) Iage Using with Gaussian Curvature and Total Variation Model 1 Deepak Kuar Gour, Sanjay Kuar Shara 1, Dept.

More information

3D Building Detection and Reconstruction from Aerial Images Using Perceptual Organization and Fast Graph Search

3D Building Detection and Reconstruction from Aerial Images Using Perceptual Organization and Fast Graph Search 436 Journal of Electrical Engineering & Technology, Vol. 3, No. 3, pp. 436~443, 008 3D Building Detection and Reconstruction fro Aerial Iages Using Perceptual Organization and Fast Graph Search Dong-Min

More information

INTRINSIC DECOMPOSITION FOR STEREOSCOPIC IMAGES

INTRINSIC DECOMPOSITION FOR STEREOSCOPIC IMAGES INTRINSIC DECOMPOSITION FOR STEREOSCOPIC IMAGES Dehua Xie 1, Shuaicheng Liu 1, Kaio Lin 2, Shuyuan Zhu 1, and Bing Zeng 1 1 School of Electronic Engineering, University of Electronic Science and Technology

More information

Automatic Image Alignment

Automatic Image Alignment Automatic Image Alignment Mike Nese with a lot of slides stolen from Steve Seitz and Rick Szeliski 15-463: Computational Photography Alexei Efros, CMU, Fall 2010 Live Homography DEMO Check out panoramio.com

More information

Homework 1. An Introduction to Neural Networks

Homework 1. An Introduction to Neural Networks Hoework An Introduction to Neural Networks -785: Introduction to Deep Learning Spring 09 OUT: January 4, 09 DUE: February 6, 09, :59 PM Start Here Collaboration policy: You are expected to coply with the

More information

Module Contact: Dr Rudy Lapeer (CMP) Copyright of the University of East Anglia Version 1

Module Contact: Dr Rudy Lapeer (CMP) Copyright of the University of East Anglia Version 1 UNIVERSITY OF EAST ANGLIA School of Coputing Sciences Main Series UG Exaination 2016-17 GRAPHICS 1 CMP-5010B Tie allowed: 2 hours Answer THREE questions. Notes are not peritted in this exaination Do not

More information

Medical Biophysics 302E/335G/ st1-07 page 1

Medical Biophysics 302E/335G/ st1-07 page 1 Medical Biophysics 302E/335G/500 20070109 st1-07 page 1 STEREOLOGICAL METHODS - CONCEPTS Upon copletion of this lesson, the student should be able to: -define the ter stereology -distinguish between quantitative

More information

Mirror Localization for a Catadioptric Imaging System by Projecting Parallel Lights

Mirror Localization for a Catadioptric Imaging System by Projecting Parallel Lights 2007 IEEE International Conference on Robotics and Autoation Roa, Italy, 10-14 April 2007 FrC2.5 Mirror Localization for a Catadioptric Iaging Syste by Projecting Parallel Lights Ryusuke Sagawa, Nobuya

More information

IMAGE MOSAICKING FOR ESTIMATING THE MOTION OF AN UNDERWATER VEHICLE. Rafael García, Xevi Cufí and Lluís Pacheco

IMAGE MOSAICKING FOR ESTIMATING THE MOTION OF AN UNDERWATER VEHICLE. Rafael García, Xevi Cufí and Lluís Pacheco IMAGE MOSAICKING FOR ESTIMATING THE MOTION OF AN UNDERWATER VEHICLE Rafael García, Xevi Cufí and Lluís Pacheco Coputer Vision and Robotics Group Institute of Inforatics and Applications, University of

More information

Ming-Wei Lee 1, Wei-Tso Lin 1, Yu-Ching Ni 2, Meei-Ling Jan 2, Yi-Chun Chen 1 * National Central University

Ming-Wei Lee 1, Wei-Tso Lin 1, Yu-Ching Ni 2, Meei-Ling Jan 2, Yi-Chun Chen 1 * National Central University Rapid Constructions of Circular-Orbit Pinhole SPECT Iaging Syste Matrices by Gaussian Interpolation Method Cobined with Geoetric Paraeter Estiations (GIMGPE Ming-Wei Lee, Wei-Tso Lin, Yu-Ching Ni, Meei-Ling

More information

An Automated Mail Sorter System using SVM Classifier

An Automated Mail Sorter System using SVM Classifier An Autoated Mail Sorter Syste using SVM Classifier Arun K.S. Assistant Professor Coputer Science & Engineering Aal Jyoti College of Engineering Kanjirappally, India Jerin Toas Assistant Professor Coputer

More information

EFFICIENT VIDEO SEARCH USING IMAGE QUERIES A. Araujo1, M. Makar2, V. Chandrasekhar3, D. Chen1, S. Tsai1, H. Chen1, R. Angst1 and B.

EFFICIENT VIDEO SEARCH USING IMAGE QUERIES A. Araujo1, M. Makar2, V. Chandrasekhar3, D. Chen1, S. Tsai1, H. Chen1, R. Angst1 and B. EFFICIENT VIDEO SEARCH USING IMAGE QUERIES A. Araujo1, M. Makar2, V. Chandrasekhar3, D. Chen1, S. Tsai1, H. Chen1, R. Angst1 and B. Girod1 1 Stanford University, USA 2 Qualco Inc., USA ABSTRACT We study

More information

3D Reconstruction for Rendering

3D Reconstruction for Rendering 3D Reconstruction for Rendering Topic in Iage-Based Modeling and Rendering CSE9 J00 Lecture 6 This lecture [DTM96] Paul E. Debevec, Caillo J. Talor, Jitendra Malik, Modeling and Rendering Architecture

More information

Image stitching. Digital Visual Effects Yung-Yu Chuang. with slides by Richard Szeliski, Steve Seitz, Matthew Brown and Vaclav Hlavac

Image stitching. Digital Visual Effects Yung-Yu Chuang. with slides by Richard Szeliski, Steve Seitz, Matthew Brown and Vaclav Hlavac Image stitching Digital Visual Effects Yung-Yu Chuang with slides by Richard Szeliski, Steve Seitz, Matthew Brown and Vaclav Hlavac Image stitching Stitching = alignment + blending geometrical registration

More information

An Integrated Processing Method for Multiple Large-scale Point-Clouds Captured from Different Viewpoints

An Integrated Processing Method for Multiple Large-scale Point-Clouds Captured from Different Viewpoints 519 An Integrated Processing Method for Multiple Large-scale Point-Clouds Captured fro Different Viewpoints Yousuke Kawauchi 1, Shin Usuki, Kenjiro T. Miura 3, Hiroshi Masuda 4 and Ichiro Tanaka 5 1 Shizuoka

More information

Preprocessing I: Within Subject John Ashburner

Preprocessing I: Within Subject John Ashburner Preprocessing I: Within Subject John Ashburner Pre-processing Overview Statistics or whatever fmri tie-series Anatoical MRI Teplate Soothed Estiate Spatial Nor Motion Correct Sooth Coregister 11 21 31

More information

Epipolar Geometry CSE P576. Dr. Matthew Brown

Epipolar Geometry CSE P576. Dr. Matthew Brown Epipolar Geometry CSE P576 Dr. Matthew Brown Epipolar Geometry Epipolar Lines, Plane Constraint Fundamental Matrix, Linear solution + RANSAC Applications: Structure from Motion, Stereo [ Szeliski 11] 2

More information

Mosaics. Today s Readings

Mosaics. Today s Readings Mosaics VR Seattle: http://www.vrseattle.com/ Full screen panoramas (cubic): http://www.panoramas.dk/ Mars: http://www.panoramas.dk/fullscreen3/f2_mars97.html Today s Readings Szeliski and Shum paper (sections

More information

EE 364B Convex Optimization An ADMM Solution to the Sparse Coding Problem. Sonia Bhaskar, Will Zou Final Project Spring 2011

EE 364B Convex Optimization An ADMM Solution to the Sparse Coding Problem. Sonia Bhaskar, Will Zou Final Project Spring 2011 EE 364B Convex Optiization An ADMM Solution to the Sparse Coding Proble Sonia Bhaskar, Will Zou Final Project Spring 20 I. INTRODUCTION For our project, we apply the ethod of the alternating direction

More information

Image stitching. Announcements. Outline. Image stitching

Image stitching. Announcements. Outline. Image stitching Announcements Image stitching Project #1 was due yesterday. Project #2 handout will be available on the web later tomorrow. I will set up a webpage for artifact voting soon. Digital Visual Effects, Spring

More information

Structure from Motion. Introduction to Computer Vision CSE 152 Lecture 10

Structure from Motion. Introduction to Computer Vision CSE 152 Lecture 10 Structure from Motion CSE 152 Lecture 10 Announcements Homework 3 is due May 9, 11:59 PM Reading: Chapter 8: Structure from Motion Optional: Multiple View Geometry in Computer Vision, 2nd edition, Hartley

More information

A wireless sensor network for visual detection and classification of intrusions

A wireless sensor network for visual detection and classification of intrusions A wireless sensor network for visual detection and classification of intrusions ANDRZEJ SLUZEK 1,3, PALANIAPPAN ANNAMALAI 2, MD SAIFUL ISLAM 1 1 School of Coputer Engineering, 2 IntelliSys Centre Nanyang

More information

Image correspondences and structure from motion

Image correspondences and structure from motion Image correspondences and structure from motion http://graphics.cs.cmu.edu/courses/15-463 15-463, 15-663, 15-862 Computational Photography Fall 2017, Lecture 20 Course announcements Homework 5 posted.

More information

Geo-activity Recommendations by using Improved Feature Combination

Geo-activity Recommendations by using Improved Feature Combination Geo-activity Recoendations by using Iproved Feature Cobination Masoud Sattari Middle East Technical University Ankara, Turkey e76326@ceng.etu.edu.tr Murat Manguoglu Middle East Technical University Ankara,

More information

Region Segmentation Region Segmentation

Region Segmentation Region Segmentation /7/ egion Segentation Lecture-7 Chapter 3, Fundaentals of Coputer Vision Alper Yilaz,, Mubarak Shah, Fall UCF egion Segentation Alper Yilaz,, Mubarak Shah, Fall UCF /7/ Laer epresentation Applications

More information

NON-RIGID OBJECT TRACKING: A PREDICTIVE VECTORIAL MODEL APPROACH

NON-RIGID OBJECT TRACKING: A PREDICTIVE VECTORIAL MODEL APPROACH NON-RIGID OBJECT TRACKING: A PREDICTIVE VECTORIAL MODEL APPROACH V. Atienza; J.M. Valiente and G. Andreu Departaento de Ingeniería de Sisteas, Coputadores y Autoática Universidad Politécnica de Valencia.

More information

Today s lecture. Image Alignment and Stitching. Readings. Motion models

Today s lecture. Image Alignment and Stitching. Readings. Motion models Today s lecture Image Alignment and Stitching Computer Vision CSE576, Spring 2005 Richard Szeliski Image alignment and stitching motion models cylindrical and spherical warping point-based alignment global

More information

The optimization design of microphone array layout for wideband noise sources

The optimization design of microphone array layout for wideband noise sources PROCEEDINGS of the 22 nd International Congress on Acoustics Acoustic Array Systes: Paper ICA2016-903 The optiization design of icrophone array layout for wideband noise sources Pengxiao Teng (a), Jun

More information

N-Views (1) Homographies and Projection

N-Views (1) Homographies and Projection CS 4495 Computer Vision N-Views (1) Homographies and Projection Aaron Bobick School of Interactive Computing Administrivia PS 2: Get SDD and Normalized Correlation working for a given windows size say

More information

Computer Vision Lecture 13

Computer Vision Lecture 13 Computer Vision Lecture 3 Recognition wit Local Features 5.2.205 astian Leibe RWTH acen ttp://www.vision.rwt-aacen.de/ leibe@vision.rwt-aacen.de Course Outline Image Processing asics Segmentation & Grouping

More information

Panoramas. Why Mosaic? Why Mosaic? Mosaics: stitching images together. Why Mosaic? Olivier Gondry. Bill Freeman Frédo Durand MIT - EECS

Panoramas. Why Mosaic? Why Mosaic? Mosaics: stitching images together. Why Mosaic? Olivier Gondry. Bill Freeman Frédo Durand MIT - EECS Olivier Gondry 6.098 Digital and Computational Photography 6.882 Advanced Computational Photography Panoramas Director of music video and commercial Special effect specialist (Morphing, rotoscoping) Today

More information

POSITION-PATCH BASED FACE HALLUCINATION VIA LOCALITY-CONSTRAINED REPRESENTATION. Junjun Jiang, Ruimin Hu, Zhen Han, Tao Lu, and Kebin Huang

POSITION-PATCH BASED FACE HALLUCINATION VIA LOCALITY-CONSTRAINED REPRESENTATION. Junjun Jiang, Ruimin Hu, Zhen Han, Tao Lu, and Kebin Huang IEEE International Conference on ultiedia and Expo POSITION-PATCH BASED FACE HALLUCINATION VIA LOCALITY-CONSTRAINED REPRESENTATION Junjun Jiang, Ruiin Hu, Zhen Han, Tao Lu, and Kebin Huang National Engineering

More information

TensorFlow and Keras-based Convolutional Neural Network in CAT Image Recognition Ang LI 1,*, Yi-xiang LI 2 and Xue-hui LI 3

TensorFlow and Keras-based Convolutional Neural Network in CAT Image Recognition Ang LI 1,*, Yi-xiang LI 2 and Xue-hui LI 3 2017 2nd International Conference on Coputational Modeling, Siulation and Applied Matheatics (CMSAM 2017) ISBN: 978-1-60595-499-8 TensorFlow and Keras-based Convolutional Neural Network in CAT Iage Recognition

More information

Image Processing for fmri John Ashburner. Wellcome Trust Centre for Neuroimaging, 12 Queen Square, London, UK.

Image Processing for fmri John Ashburner. Wellcome Trust Centre for Neuroimaging, 12 Queen Square, London, UK. Iage Processing for fmri John Ashburner Wellcoe Trust Centre for Neuroiaging, 12 Queen Square, London, UK. Contents * Preliinaries * Rigid-Body and Affine Transforations * Optiisation and Objective Functions

More information

Relief shape inheritance and graphical editor for the landscape design

Relief shape inheritance and graphical editor for the landscape design Relief shape inheritance and graphical editor for the landscape design Egor A. Yusov Vadi E. Turlapov Nizhny Novgorod State University after N. I. Lobachevsky Nizhny Novgorod Russia yusov_egor@ail.ru vadi.turlapov@cs.vk.unn.ru

More information

Detection of Outliers and Reduction of their Undesirable Effects for Improving the Accuracy of K-means Clustering Algorithm

Detection of Outliers and Reduction of their Undesirable Effects for Improving the Accuracy of K-means Clustering Algorithm Detection of Outliers and Reduction of their Undesirable Effects for Iproving the Accuracy of K-eans Clustering Algorith Bahan Askari Departent of Coputer Science and Research Branch, Islaic Azad University,

More information

Step-by-Step Model Buidling

Step-by-Step Model Buidling Step-by-Step Model Buidling Review Feature selection Feature selection Feature correspondence Camera Calibration Euclidean Reconstruction Landing Augmented Reality Vision Based Control Sparse Structure

More information

Camera Geometry II. COS 429 Princeton University

Camera Geometry II. COS 429 Princeton University Camera Geometry II COS 429 Princeton University Outline Projective geometry Vanishing points Application: camera calibration Application: single-view metrology Epipolar geometry Application: stereo correspondence

More information

The Internal Conflict of a Belief Function

The Internal Conflict of a Belief Function The Internal Conflict of a Belief Function Johan Schubert Abstract In this paper we define and derive an internal conflict of a belief function We decopose the belief function in question into a set of

More information

6.098 Digital and Computational Photography Advanced Computational Photography. Panoramas. Bill Freeman Frédo Durand MIT - EECS

6.098 Digital and Computational Photography Advanced Computational Photography. Panoramas. Bill Freeman Frédo Durand MIT - EECS 6.098 Digital and Computational Photography 6.882 Advanced Computational Photography Panoramas Bill Freeman Frédo Durand MIT - EECS Lots of slides stolen from Alyosha Efros, who stole them from Steve Seitz

More information

CSCI 5980/8980: Assignment #4. Fundamental Matrix

CSCI 5980/8980: Assignment #4. Fundamental Matrix Submission CSCI 598/898: Assignment #4 Assignment due: March 23 Individual assignment. Write-up submission format: a single PDF up to 5 pages (more than 5 page assignment will be automatically returned.).

More information

Image Warping and Mosacing

Image Warping and Mosacing Image Warping and Mosacing 15-463: Rendering and Image Processing Alexei Efros with a lot of slides stolen from Steve Seitz and Rick Szeliski Today Mosacs Image Warping Homographies Programming Assignment

More information

Automatic Image Alignment (feature-based)

Automatic Image Alignment (feature-based) Automatic Image Alignment (feature-based) Mike Nese with a lot of slides stolen from Steve Seitz and Rick Szeliski 15-463: Computational Photography Alexei Efros, CMU, Fall 2006 Today s lecture Feature

More information

Structure from Motion

Structure from Motion 11/18/11 Structure from Motion Computer Vision CS 143, Brown James Hays Many slides adapted from Derek Hoiem, Lana Lazebnik, Silvio Saverese, Steve Seitz, and Martial Hebert This class: structure from

More information

Instance-level recognition

Instance-level recognition Instance-level recognition 1) Local invariant features 2) Matching and recognition with local features 3) Efficient visual search 4) Very large scale indexing Matching of descriptors Matching and 3D reconstruction

More information

Novel Image Representation and Description Technique using Density Histogram of Feature Points

Novel Image Representation and Description Technique using Density Histogram of Feature Points Novel Iage Representation and Description Technique using Density Histogra of Feature Points Keneilwe ZUVA Departent of Coputer Science, University of Botswana, P/Bag 00704 UB, Gaborone, Botswana and Tranos

More information

Transformations. Prof. George Wolberg Dept. of Computer Science City College of New York

Transformations. Prof. George Wolberg Dept. of Computer Science City College of New York Transforations Prof. George Wolberg Dept. of Coputer Science City College of New York Objectives Introduce standard transforations - Rotations - Translation - Scaling - Shear Derive hoogeneous coordinate

More information

Computational Optical Imaging - Optique Numerique. -- Multiple View Geometry and Stereo --

Computational Optical Imaging - Optique Numerique. -- Multiple View Geometry and Stereo -- Computational Optical Imaging - Optique Numerique -- Multiple View Geometry and Stereo -- Winter 2013 Ivo Ihrke with slides by Thorsten Thormaehlen Feature Detection and Matching Wide-Baseline-Matching

More information

Single-view Metrology and Camera Calibration

Single-view Metrology and Camera Calibration Sigle-iew Metrology ad Caera Calibratio Coputer Visio Jia-Bi Huag, Virgiia Tech May slides fro S. Seitz ad D. Hoie Adiistratie stuffs HW 2 due :59 PM o Oct 9 th Ask/discuss questios o Piazza Office hour

More information

Cassia County School District #151. Expected Performance Assessment Students will: Instructional Strategies. Performance Standards

Cassia County School District #151. Expected Performance Assessment Students will: Instructional Strategies. Performance Standards Unit 1 Congruence, Proof, and Constructions Doain: Congruence (CO) Essential Question: How do properties of congruence help define and prove geoetric relationships? Matheatical Practices: 1. Make sense

More information

ELEN : Project Progress Report. Second Order Delay Computation for RC networks with Non-Tree Topology

ELEN : Project Progress Report. Second Order Delay Computation for RC networks with Non-Tree Topology ELEN 689-603: Project Progress eport Second Order Delay Coputation for C networks wi Non-Tree Topology ajeshwary Tayade 000 07 573 //003 PDF created wi pdffactory trial version www.pdffactory.co : Introduction

More information

Camera Calibration. COS 429 Princeton University

Camera Calibration. COS 429 Princeton University Camera Calibration COS 429 Princeton University Point Correspondences What can you figure out from point correspondences? Noah Snavely Point Correspondences X 1 X 4 X 3 X 2 X 5 X 6 X 7 p 1,1 p 1,2 p 1,3

More information

Last lecture. Passive Stereo Spacetime Stereo

Last lecture. Passive Stereo Spacetime Stereo Last lecture Passive Stereo Spacetime Stereo Today Structure from Motion: Given pixel correspondences, how to compute 3D structure and camera motion? Slides stolen from Prof Yungyu Chuang Epipolar geometry

More information

Identifying Converging Pairs of Nodes on a Budget

Identifying Converging Pairs of Nodes on a Budget Identifying Converging Pairs of Nodes on a Budget Konstantina Lazaridou Departent of Inforatics Aristotle University, Thessaloniki, Greece konlaznik@csd.auth.gr Evaggelia Pitoura Coputer Science and Engineering

More information

Optimized stereo reconstruction of free-form space curves based on a nonuniform rational B-spline model

Optimized stereo reconstruction of free-form space curves based on a nonuniform rational B-spline model 1746 J. Opt. Soc. A. A/ Vol. 22, No. 9/ Septeber 2005 Y. J. Xiao and Y. F. Li Optiized stereo reconstruction of free-for space curves based on a nonunifor rational B-spline odel Yi Jun Xiao Departent of

More information

GROUND VEHICLE ATTITUDE ESTIMATION THROUGH MAGNETIC-INERTIAL SENSOR FUSION

GROUND VEHICLE ATTITUDE ESTIMATION THROUGH MAGNETIC-INERTIAL SENSOR FUSION F16-VDCF-4 GROUND VEHICLE ATTITUDE ESTIMATION THROUGH MAGNETIC-INERTIAL SENSOR FUSION 1 Hwang, Yoonjin * ; 1 Seibu, Choi 1 Korea Advanced Institute of Science and Technology, S. Korea KEYWORDS Vehicle

More information

A Novel Fast Constructive Algorithm for Neural Classifier

A Novel Fast Constructive Algorithm for Neural Classifier A Novel Fast Constructive Algorith for Neural Classifier Xudong Jiang Centre for Signal Processing, School of Electrical and Electronic Engineering Nanyang Technological University Nanyang Avenue, Singapore

More information

Preprocessing of fmri data (basic)

Preprocessing of fmri data (basic) Preprocessing of fmri data (basic) Practical session SPM Course 2016, Zurich Andreea Diaconescu, Maya Schneebeli, Jakob Heinzle, Lars Kasper, and Jakob Sieerkus Translational Neuroodeling Unit (TNU) Institute

More information

Data Acquisition of Obstacle Shapes for Fish Robots

Data Acquisition of Obstacle Shapes for Fish Robots Proceedings of the 2nd WEA International Conference on Dynaical ystes and Control, Bucharest, oania, October -17, 6 Data Acquisition of Obstacle hapes for Fish obots EUNG Y. NA, DAEJUNG HIN, JIN Y. KIM,

More information

Feature-Centric Evaluation for Efficient Cascaded Object Detection

Feature-Centric Evaluation for Efficient Cascaded Object Detection IEEE Conference on Coputer Vision and Pattern (CVPR, 2004 Feature-Centric Evaluation for Efficient Cascaded Object Detection Henry Schneideran 1 Robotics Institute Carnegie Mellon University Pittsburgh,

More information

Local features: detection and description. Local invariant features

Local features: detection and description. Local invariant features Local features: detection and description Local invariant features Detection of interest points Harris corner detection Scale invariant blob detection: LoG Description of local patches SIFT : Histograms

More information

Using Imperialist Competitive Algorithm in Optimization of Nonlinear Multiple Responses

Using Imperialist Competitive Algorithm in Optimization of Nonlinear Multiple Responses International Journal of Industrial Engineering & Production Research Septeber 03, Volue 4, Nuber 3 pp. 9-35 ISSN: 008-4889 http://ijiepr.iust.ac.ir/ Using Iperialist Copetitive Algorith in Optiization

More information

Instance-level recognition

Instance-level recognition Instance-level recognition 1) Local invariant features 2) Matching and recognition with local features 3) Efficient visual search 4) Very large scale indexing Matching of descriptors Matching and 3D reconstruction

More information

Feature Matching and Robust Fitting

Feature Matching and Robust Fitting Feature Matching and Robust Fitting Computer Vision CS 143, Brown Read Szeliski 4.1 James Hays Acknowledgment: Many slides from Derek Hoiem and Grauman&Leibe 2008 AAAI Tutorial Project 2 questions? This

More information

Multiview Stereo COSC450. Lecture 8

Multiview Stereo COSC450. Lecture 8 Multiview Stereo COSC450 Lecture 8 Stereo Vision So Far Stereo and epipolar geometry Fundamental matrix captures geometry 8-point algorithm Essential matrix with calibrated cameras 5-point algorithm Intersect

More information

Solving the Damage Localization Problem in Structural Health Monitoring Using Techniques in Pattern Classification

Solving the Damage Localization Problem in Structural Health Monitoring Using Techniques in Pattern Classification Solving the Daage Localization Proble in Structural Health Monitoring Using Techniques in Pattern Classification CS 9 Final Project Due Dec. 4, 007 Hae Young Noh, Allen Cheung, Daxia Ge Introduction Structural

More information

Stereo and Epipolar geometry

Stereo and Epipolar geometry Previously Image Primitives (feature points, lines, contours) Today: Stereo and Epipolar geometry How to match primitives between two (multiple) views) Goals: 3D reconstruction, recognition Jana Kosecka

More information

3 Conference on Inforation Sciences and Systes, The Johns Hopkins University, March, 3 Sensitivity Characteristics of Cross-Correlation Distance Metric and Model Function F. Porikli Mitsubishi Electric

More information

Camera Calibration. Schedule. Jesus J Caban. Note: You have until next Monday to let me know. ! Today:! Camera calibration

Camera Calibration. Schedule. Jesus J Caban. Note: You have until next Monday to let me know. ! Today:! Camera calibration Camera Calibration Jesus J Caban Schedule! Today:! Camera calibration! Wednesday:! Lecture: Motion & Optical Flow! Monday:! Lecture: Medical Imaging! Final presentations:! Nov 29 th : W. Griffin! Dec 1

More information

Practical Algorithms of Geometrical Migration

Practical Algorithms of Geometrical Migration XIII International Conference on Ground penetrating adar, Lecce (Italy), June. ractical Algorits of Geoetrical Migration F.D. Edesky,.A. Morozov, D.E. Edesky, A.V. opov Dept. of radio wave propagation,

More information

Structure from motion

Structure from motion Structure from motion Structure from motion Given a set of corresponding points in two or more images, compute the camera parameters and the 3D point coordinates?? R 1,t 1 R 2,t 2 R 3,t 3 Camera 1 Camera

More information

CSE 527: Introduction to Computer Vision

CSE 527: Introduction to Computer Vision CSE 527: Introduction to Computer Vision Week 5 - Class 1: Matching, Stitching, Registration September 26th, 2017 ??? Recap Today Feature Matching Image Alignment Panoramas HW2! Feature Matches Feature

More information

Reminder: Lecture 20: The Eight-Point Algorithm. Essential/Fundamental Matrix. E/F Matrix Summary. Computing F. Computing F from Point Matches

Reminder: Lecture 20: The Eight-Point Algorithm. Essential/Fundamental Matrix. E/F Matrix Summary. Computing F. Computing F from Point Matches Reminder: Lecture 20: The Eight-Point Algorithm F = -0.00310695-0.0025646 2.96584-0.028094-0.00771621 56.3813 13.1905-29.2007-9999.79 Readings T&V 7.3 and 7.4 Essential/Fundamental Matrix E/F Matrix Summary

More information

HIGH PERFORMANCE PRE-SEGMENTATION ALGORITHM FOR SONAR IMAGES

HIGH PERFORMANCE PRE-SEGMENTATION ALGORITHM FOR SONAR IMAGES HIGH PERFORMANCE PRE-SEGMENTATION ALGORITHM FOR SONAR IMAGES Benjain Lehann*, Konstantinos Siantidis*, Dieter Kraus** *ATLAS ELEKTRONIK GbH Sebaldsbrücker Heerstraße 235 D-28309 Breen, GERMANY Eail: benjain.lehann@atlas-elektronik.co

More information

Camera models and calibration

Camera models and calibration Camera models and calibration Read tutorial chapter 2 and 3. http://www.cs.unc.edu/~marc/tutorial/ Szeliski s book pp.29-73 Schedule (tentative) 2 # date topic Sep.8 Introduction and geometry 2 Sep.25

More information