Lecture 6: Multiple-View Geometry: Stereo, Motion, Features

Structure from Stereo

two or more images, calibrated cameras (K(f), R, T known)
http://en.wikipedia.org/wiki/Stereo_vision

ref.
UCR stereographs
http://www.cmp.ucr.edu/site/exhibitions/stereo/
The Art of Stereo Photography
http://www.photostuff.co.uk/stereo.htm
History of Stereo Photography
http://www.rpi.edu/~ruiz/stereo_history/text/historystereog.html
Double Exposure
http://home.centurytel.net/s3dcor/index.html
Stereo Photography
http://www.shortcourses.com/book01/chapter09.htm
3D Photography links
http://www.studyweb.com/links/5243.html
National Stereoscopic Association
http://204.248.144.203/3dLibrary/welcome.html
Books on Stereo Photography
http://userwww.sfsu.edu/~hl/3d.biblio.html

http://en.wikipedia.org/wiki/Epipolar_geometry

http://en.wikipedia.org/wiki/Image_rectification
http://en.wikipedia.org/wiki/Rectification_(geometry)


Computing Rectifying Homographies for Stereo Vision
Charles Loop, Zhengyou Zhang

ComputingRectifyingHomographies.pdf

Microsoft Research, April 8, 1999 


cross correlation
http://en.wikipedia.org/wiki/Cross_correlation



S. M. Seitz and C. R. Dyer, View Morphing, Proc. SIGGRAPH 96, 1996, pp. 21-30.

L. McMillan and G. Bishop. Plenoptic Modeling: An Image-Based Rendering System, Proc. of SIGGRAPH 95, 1995, pp. 39-46.

> Stereo Reconstruction
1) Calibrate cameras
2) Rectify images
3) Compute disparity
4) Estimate depth

SSD (Sum of Squared Difference) to choose the baseline 

Structure from Motion

Incremental Motion: Kinematic relation  

7 Unknowns respect to motion
 translation:  V = (VX , VY , VZ )
 rotation :     W = (WX, WY, WZ )
 depth :         Z
2 equations for one point
 
6 points with 12 unknowns -> 12 equations

 
http://en.wikipedia.org/wiki/Structure_from_motion
http://en.wikipedia.org/wiki/Motion_field

http://en.wikipedia.org/wiki/Nodal_point#Nodal_points


optical flow, image flow
http://en.wikipedia.org/wiki/Optical_flow

Horn-Schunck Algorithm
http://en.wikipedia.org/wiki/Horn-Schunck_algorithm

Determining Optical Flow

horn_schunck_DeterminingOpticalFlow.pdf

Berthold K.P. Horn and Brian G. Schunck
Artificial Intelligence Laboratory, Massachusetts Institute of Technology



Larger Motion - feature matching

Lucas Kanade style motion estimation
http://en.wikipedia.org/wiki/Lucas%E2%80%93Kanade_method


Good Features to Track

shi_tomasi_GoodFeaturesToTrack.pdf

Jianbo Shi (Computer Science Department, Cornel1 University)
Carlo Tomasi (Computer Science Department, Stanford University)


http://en.wikipedia.org/wiki/Kalman_filter