Integrating Algebraic Functions of Views with Indexing and Learning for 3D Object Recognition

Integrating Algebraic Functions of Views with Indexing and Learning for 3D Object Recognition (2004)

Cached

Download Links

by Wenjing Li , George Bebis , Nikolaos Bourbakis

Venue:	in Proc. of IEEE Workshop on Learning in Computer Vision and Pattern Recognition
Citations:	3 - 2 self

Versions

Version History

Metadata Version 3

User correction supplied by SystemCorrections

Datum	Value	Source
TITLE	Integrating Algebraic Functions of Views with Indexing and Learning for 3D Object Recognition	user correction - Legacy Corrections
AUTHOR NAME	Wenjing Li	user correction - Legacy Corrections
AUTHOR AFFIL	Computer Vision Laboratory; University of Nevada, Reno	user correction - Legacy Corrections
AUTHOR NAME	George Bebis	user correction - Legacy Corrections
AUTHOR AFFIL	Computer Vision Laboratory; University of Nevada, Reno	user correction - Legacy Corrections
AUTHOR NAME	Nikolaos Bourbakis	user correction - Legacy Corrections
ABSTRACT	This paper focuses on the problem of 3D object recognition from di#erent viewing angles and positions. In particular, we propose a new approach that integrates Algebraic Functions of View (AFoVs) with indexing and learning. During training, we consider groups of point features and we represent a sparse number of views that they can produce in a k-d tree. Moreover, we learn the manifold formed by a dense number of views using mixture models and Expectation-Maximization (EM). Learning takes place in a "universal", lower-dimensional, space computed through Random Projection (RP). The images that a group of model points can produce are computed o#-line using AFoVs by combining a small number of reference views that contain the group. Rigidity constraints are imposed during this step to remove unrealistic views. During recognition, groups of point features are extracted from the scene and used to retrieve from the k-d tree the most feasible model groups that might have produced them. To reduce the number of hypotheses for verification, we rank them by computing the probability that the hypothesized model group is present in the scene. Only hypotheses ranked high enough are considered for further verification. The proposed method has been evaluated using both artificial and real data, illustrating good performance.	user correction - Legacy Corrections
YEAR	2004	INFERENCE
VENUE	in Proc. of IEEE Workshop on Learning in Computer Vision and Pattern Recognition	INFERENCE
VENUE TYPE	CONFERENCE	INFERENCE
CITATIONS	27 found	ParsCit 1.0