Slightly Supervised Learning of Part-Based Appearance Models

Authors:
Lexing Xie;Patrick Perez
Affiliations:
Columbia University, New York, NY;Irisa/Inria, Cedex
Venue:
CVPRW '04 Proceedings of the 2004 Conference on Computer Vision and Pattern Recognition Workshop (CVPRW'04) Volume 6 - Volume 06
Year:
2004

Citing 0
Cited 6

Low-cost distributed learning of a Gaussian mixture model for multimedia content-based indexing on a peer-to-peer network

Proceedings of the 7th ACM SIGMM international workshop on Multimedia information retrieval
A Discriminative Learning Framework with Pairwise Constraints for Video Object Classification

IEEE Transactions on Pattern Analysis and Machine Intelligence
Object recognition using proportion-based prior information: Application to fisheries acoustics

Pattern Recognition Letters
Object recognition via local patch labelling

Proceedings of the First international conference on Deterministic and Statistical Methods in Machine Learning
A generative model for multi class object recognition and detection

TAINN'05 Proceedings of the 14th Turkish conference on Artificial Intelligence and Neural Networks
Variational learning of a Dirichlet process of generalized Dirichlet distributions for simultaneous clustering and feature selection

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

We extend the GMM-based approach of [Selection of scale-invariant parts for object class recognition], for learning part-based appearance models of object categories, to the unsupervised case where positive examples are corrupted with clutter. To this end, we derive an original version of EM which is able to fit one GMM per class based on partially labeled data. We also allow ourselves a small fraction of un-corrupted positive examples, thus obtaining an effective, yet cheap, slightly supervised learning. Proposed technique allows as well a saliency-based ranking and selection of learnt mixture components. Experiments show that both the semi-supervised GMM fitting with side information and the component selection are effective in identifying salient patches in the appearance of a class of objects. They are thus promising tools to learn class-specific models and detectors similar to those by Weber et al.[Unsupervised learning of models for recognition],but at a lower computational cost, while accommodating larger numbers of atomic parts.