Supervised hierarchical Pitman-Yor process for natural scene segmentation

Authors:
A. Shyr;T. Darrell;M. Jordan;R. Urtasun
Affiliations:
UC Berkeley, Berkeley, CA, USA;ICSI, UC Berkeley, Berkeley, CA, USA;UC Berkeley, Berkeley, CA, USA;TTI Chicago, Chicago, IL, USA
Venue:
CVPR '11 Proceedings of the 2011 IEEE Conference on Computer Vision and Pattern Recognition
Year:
2011

Citing 0
Cited 1

A Markov random field-regulated Pitman-Yor process prior for spatially constrained data clustering

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

From conventional wisdom and empirical studies of annotated data, it has been shown that visual statistics such as object frequencies and segment sizes follow power law distributions. Previous work has shown that both kinds of power-law behavior can be captured by using a hierarchical Pitman-Yor process prior within a nonparametric Bayesian approach to scene segmentation. In this paper, we add label information into the previously unsupervised model. Our approach exploits the labelled data by adding constraints on the parameter space during the variational learning phase. We evaluate our formulation on the LabelMe natural scene dataset, and show the effectiveness of our approach.