Local Image Features Resulting from 3-Dimensional Geometric Features, Illumination, and Movement: I

Authors:
James Damon;Peter Giblin;Gareth Haslinger
Affiliations:
Department of Mathematics, University of North Carolina, Chapel Hill, USA 27599;Department of Mathematical Sciences, University of Liverpool, Liverpool, UK L69 7ZL;Department of Mathematical Sciences, University of Liverpool, Liverpool, UK L69 7ZL
Venue:
International Journal of Computer Vision
Year:
2009

Citing 8
Cited 1

Machine interpretation of line drawings

Machine interpretation of line drawings
On the classification of views of piecewise smooth objects

Image and Vision Computing - Special issue: papers from the second Alvey Vision Conference
Solid shape

Solid shape
Shape from shading

Shape from shading
Computing exact aspect graphs of curved objects: algebraic surfaces

International Journal of Computer Vision
Topographic Maps and Local Contrast Changes in Natural Images

International Journal of Computer Vision
Robot Vision

Robot Vision
Interpreting Line Drawings of Objects with K-Vertices

GMP '04 Proceedings of the Geometric Modeling and Processing 2004

Local Image Features Resulting from 3-Dimensional Geometric Features, Illumination, and Movement: II

SIAM Journal on Imaging Sciences

Quantified Score

Hi-index	0.00

Visualization

Abstract

We study images of smooth or piecewise smooth objects illuminated by a single light source, with only background illumination from other sources. The objects may have geometric features (F), namely surface markings, boundary edges, creases and corners; and shade features (S), namely shade curves and cast shadow curves. We determine the local stable interactions between these and apparent contours (C) for the various configurations of F, S, C, and we concisely summarize them using an "alphabet" of local curve configurations. We further determine the generic transitions for the configurations resulting from viewer movement. These classifications are obtained using the methods of singularity theory, which allows us to ensure that our lists are complete, in some cases correcting earlier attempts at similar classifications.