Title: TODAY: Affine Structure from Motion IV
1TODAY Affine Structure from Motion IV
- Affine Structure from Motion from two Views
- Affine Epipolar Geometry
- An Algebraic Approach
- Affine Structure from Motion from two Views
- An SVD-Based Approach Factorization
- Computing Euclidean Upgrades
- Reading Projective Structure from Motion
- This file on feather.ai.uiuc.edu in
pub/ponce/lect19.ppt.gz - Programming assignment on feather in
pub/ponce/prog19.tex, - due next Thursday.
2The Affine Epipolar Constraint
Note the epipolar lines are parallel.
3Affine Epipolar Geometry
4The Affine Fundamental Matrix
where
5An Affine Trick..
Algebraic Scene Reconstruction Method
6The Affine Structure of Affine Images
Suppose we observe a scene with m fixed cameras..
The set of all images of a fixed scene is a 3D
affine space!
7has rank 4!
8From Affine to Vectorial Structure
Idea pick one of the points (or their center of
mass) as the origin.
9What if we could factorize D? (Tomasi and
Kanade, 1992)
Affine SFM is solved!
Singular Value Decomposition
We can take
10From uncalibrated to calibrated cameras
Weak-perspective camera
Calibrated camera
Problem what is Q ?
Note Absolute scale cannot be recovered. The
Euclidean shape (defined up to an arbitrary
similitude) is recovered.
11(No Transcript)
12The Projective Structure-from-Motion Problem
Given m perspective images of n fixed points P
we can write
j
2mn equations in 11m3n unknowns
Overconstrained problem, that can be solved using
(non-linear) least squares!
13The Projective Ambiguity of Projective SFM
When the intrinsic and extrinsic parameters are
unknown
and Q is an arbitrary non-singular 4x4 matrix.
Q is a projective transformation.
14Projective Spaces (Semi-Formal) Definition
153
A Model of P( R )