The Specialized Mappings Architecture


Autoria(s): Rosales, Rómer; Sclaroff, Stan
Data(s)

20/10/2011

20/10/2011

28/03/2003

Resumo

A probabilistic, nonlinear supervised learning model is proposed: the Specialized Mappings Architecture (SMA). The SMA employs a set of several forward mapping functions that are estimated automatically from training data. Each specialized function maps certain domains of the input space (e.g., image features) onto the output space (e.g., articulated body parameters). The SMA can model ambiguous, one-to-many mappings that may yield multiple valid output hypotheses. Once learned, the mapping functions generate a set of output hypotheses for a given input via a statistical inference procedure. The SMA inference procedure incorporates an inverse mapping or feedback function in evaluating the likelihood of each of the hypothesis. Possible feedback functions include computer graphics rendering routines that can generate images for given hypotheses. The SMA employs a variant of the Expectation-Maximization algorithm for simultaneous learning of the specialized domains along with the mapping functions, and approximate strategies for inference. The framework is demonstrated in a computer vision system that can estimate the articulated pose parameters of a human’s body or hands, given silhouettes from a single image. The accuracy and stability of the SMA are also tested using synthetic images of human bodies and hands, where ground truth is known.

U.S. Office of Naval Research (N000140310108, N000140110444); U.S.National Science Foundation (IIS-0208876, IIS-9809340)

Identificador

http://hdl.handle.net/2144/1503

Idioma(s)

en_US

Publicador

Boston University Computer Science Department

Relação

BUCS Technical Reports;BUCS-TR-2003-007

Palavras-Chave #Supervised learning #Statistical inference #Mixture models #Expectation maximization algorithm #Articulated structure estimation #Human body pose #Hand shape
Tipo

Technical Report