Probabilistic pose recovery using learned hierarchical object models

Detry, R., Pugeault, N. and Piater, J. (2008) Probabilistic pose recovery using learned hierarchical object models. In: Caputo, B. and Vincze, M. (eds.) Cognitive Vision: 4th International Workshop, ICVW 2008, Santorini, Greece, May 12, 2008, Revised Selected Papers. Series: Lecture notes in computer science (5329). Springer: Berlin ; New York, pp. 107-120. ISBN 9783540927808 (doi: 10.1007/978-3-540-92781-5_9)

Full text not currently available from Enlighten.

Abstract

This paper presents a probabilistic representation for 3D objects, and details the mechanism of inferring the pose of real-world objects from vision. Our object model has the form of a hierarchy of increasingly expressive 3D features, and probabilistically represents 3D relations between these. Features at the bottom of the hierarchy are bound to local perceptions; while we currently only use visual features, our method can in principle incorporate features from diverse modalities within a coherent framework. Model instances are detected using a Nonparametric Belief Propagation algorithm which propagates evidence through the hierarchy to infer globally consistent poses for every feature of the model. Belief updates are managed by an importance-sampling mechanism that is critical for efficient and precise propagation. We conclude with a series of pose estimation experiments on real objects, along with quantitative performance evaluation.

Item Type:Book Sections
Status:Published
Glasgow Author(s) Enlighten ID:Pugeault, Dr Nicolas
Authors: Detry, R., Pugeault, N., and Piater, J.
College/School:College of Science and Engineering > School of Computing Science
Publisher:Springer
ISBN:9783540927808

University Staff: Request a correction | Enlighten Editors: Update this record