MetaCartSign in to MyCiteSeer

Include Citations | Advanced Search | Help

Include Citations | Advanced Search | Help

  Shape from Symmetry- Detecting and Exploiting Symmetry in Affine Images

Download:
pdf | ps
by Dipti Prasad, Mukherjee Andrew Zisserman, Michael Brady
http://www.robots.ox.ac.uk/~vgg/vggpapers/Mukherjee94.ps.gz
Add To MetaCart

Abstract:

We investigate the constraints placed on the image projection of a planar object having local reflectional symmetry. Under the affine approximation to projection, we demonstrate an efficient (low complexity) algorithm for detecting and verifying symmetries despite the distorting effects of image skewing. The symmetries are utilised for three distinct tasks: Firstly, determining image backprojection up to a similarity transformation ambiguity; Secondly, determining the object plane orientation (slant and tilt); Thirdly, as a test for non co-planarity amongst a collection of objects. These results are illustrated throughout with examples from

Citations

358 Perceptual Organization and Visual Recognition – Lowe - 1985
298 Object Recognition by Computer: The Role of Geometric Constraints – Grimson - 1990
234 An efficient and accurate camera calibration technique for 3D machine vision – Tsai - 1986
132 Shape description using weighted symmetric axis features – Blum, Nagel - 1978
96 Object recognition by affine invariant matching – Lamdan, Schwartz, et al. - 1988
75 Smoothed local symmetries and their implementation – Brady, Asada - 1984
73 Recovery of the three-dimensional shape of an object from a single view – Kanade - 1980
56 Shape from texture: estimation, isotropy and moments – Blake, Marinos - 1989
50 Canonical Frames for Planar Object Recognition – Rothwell, Forsyth
39 Geometry and analysis of projective spaces – Springer - 1964
29 Efficient model library access by projectively invariant indexing functions – Rothwell, Zisserman, et al. - 1992
28 An extremum principle for shape from contour – Brady, Yuille - 1984
27 Semidifferential invariants – Gool, Moons, et al. - 1992
14 On the geometric interpretation of image contours – Horaud, Brady - 1987
14 Relative motion and pose from arbitrary plane curves – Rothwell, Zisserman, et al. - 1992
13 Local symmetry of plane curves – Giblin, Brassett - 1985
11 Local Rotational Symmetries – Fleck - 1986
10 Finding Edges and – Canny - 1983
10 Using a mixed wave/diffusion process to elicit the symmetry set – Scott, Turner, et al. - 1989
9 Recognition of disoriented shapes – Corballis - 1988
4 Recognition-by-components: a theory of human image understanding – Beiderman - 1987
4 Generating and generalising models of visual objects – Connell, Brady - 1987
4 Symmetry in Chaos: a search for pattern in mathematics, art and nature – Field, Golubitsky - 1992
3 Recognising Parameterized Models from Range Data – Reid - 1991
3 Recognition of fish species by colour and shape – Strachan - 1993
3 Skewed symmetry: A nonaccidental property used to perceive visual forms – Wagemans - 1993
2 Hierarchical decomposition and axial representation of shape – Rom - 1993
1 Extracting Structure from Single Affine Views of 3D Point Sets with One or Two Bilateral Symmetries – Fawcett, Zisserman - 1993
1 A Proof of Theorem 1 Suppose two image curves fl and fl are images of two sides of a planar object with bilateral symmetry, and image projection can be represented by an affine transformation. Then the transformation between fl and fl has the following pr – Weyl - 1952
1 fl are related by an affine transformation. That is, if x is a point on fl then there is a point x on fl such that: x = Ax + b where A is a non-singular 2 \Theta 2 matrix, and b is a two-vector. Proof The first statement is a straightforward consequence o – fl
1 fl are related by a projective transformation. That is, if x is a point on fl then there is a point x on fl such that: x = Tx (23) where T is a non-singular 3 \Theta 3 matrix, and x and x are homogeneous three vectors – fl