TY - GEN

T1 - An efficient PAC algorithm for reconstructing a mixture of lines

AU - Dasgupta, Sanjoy

AU - Pavlov, Elan

AU - Singer, Yoram

N1 - Publisher Copyright:
© Springer-Verlag Berlin Heidelberg 2002.

PY - 2002

Y1 - 2002

N2 - In this paper we study the learnability of a mixture of lines model which is of great importance in machine vision, computer graphics, and computer aided design applications. The mixture of lines is a partially-probabilistic model for an image composed of line-segments. Observations are generated by choosing one of the lines at random and picking a point at random from the chosen line. Each point is contaminated with some noise whose distribution is unknown, but which is bounded in magnitude. Our goal is to discover efficiently and rather accurately the line-segments that generated the noisy observations. We describe and analyze an efficient probably approximately correct (PAC) algorithm for solving the problem. Our algorithm combines techniques from planar geometry with simple large deviation tools and is simple to implement.

AB - In this paper we study the learnability of a mixture of lines model which is of great importance in machine vision, computer graphics, and computer aided design applications. The mixture of lines is a partially-probabilistic model for an image composed of line-segments. Observations are generated by choosing one of the lines at random and picking a point at random from the chosen line. Each point is contaminated with some noise whose distribution is unknown, but which is bounded in magnitude. Our goal is to discover efficiently and rather accurately the line-segments that generated the noisy observations. We describe and analyze an efficient probably approximately correct (PAC) algorithm for solving the problem. Our algorithm combines techniques from planar geometry with simple large deviation tools and is simple to implement.

UR - http://www.scopus.com/inward/record.url?scp=84942749108&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84942749108&partnerID=8YFLogxK

U2 - 10.1007/3-540-36169-3_28

DO - 10.1007/3-540-36169-3_28

M3 - Conference contribution

AN - SCOPUS:84942749108

SN - 3540001700

SN - 9783540001706

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 351

EP - 364

BT - Algorithmic Learning Theory - 13th International Conference, ALT 2002, Proceedings

A2 - Cesa-Bianchi, Nicolò

A2 - Numao, Masayuki

A2 - Reischuk, Rüdiger

PB - Springer Verlag

T2 - 13th International Conference on Algorithmic Learning Theory, ALT 2002

Y2 - 24 November 2002 through 26 November 2002

ER -