Saved in:
Bibliographic Details
Main Authors: Baltaxe, Michael, Meer, Peter, Lindenbaum, Michael
Format: Preprint
Published: 2015
Subjects:
Online Access:https://arxiv.org/abs/1504.06507
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913002277568512
author Baltaxe, Michael
Meer, Peter
Lindenbaum, Michael
author_facet Baltaxe, Michael
Meer, Peter
Lindenbaum, Michael
contents The goal of image oversegmentation is to divide an image into several pieces, each of which should ideally be part of an object. One of the simplest and yet most effective oversegmentation algorithms is known as local variation (LV) (Felzenszwalb and Huttenlocher 2004). In this work, we study this algorithm and show that algorithms similar to LV can be devised by applying different statistical models and decisions, thus providing further theoretical justification and a well-founded explanation for the unexpected high performance of the LV approach. Some of these algorithms are based on statistics of natural images and on a hypothesis testing decision; we denote these algorithms probabilistic local variation (pLV). The best pLV algorithm, which relies on censored estimation, presents state-of-the-art results while keeping the same computational complexity of the LV algorithm.
format Preprint
id arxiv_https___arxiv_org_abs_1504_06507
institution arXiv
publishDate 2015
record_format arxiv
spellingShingle Local Variation as a Statistical Hypothesis Test
Baltaxe, Michael
Meer, Peter
Lindenbaum, Michael
Computer Vision and Pattern Recognition
The goal of image oversegmentation is to divide an image into several pieces, each of which should ideally be part of an object. One of the simplest and yet most effective oversegmentation algorithms is known as local variation (LV) (Felzenszwalb and Huttenlocher 2004). In this work, we study this algorithm and show that algorithms similar to LV can be devised by applying different statistical models and decisions, thus providing further theoretical justification and a well-founded explanation for the unexpected high performance of the LV approach. Some of these algorithms are based on statistics of natural images and on a hypothesis testing decision; we denote these algorithms probabilistic local variation (pLV). The best pLV algorithm, which relies on censored estimation, presents state-of-the-art results while keeping the same computational complexity of the LV algorithm.
title Local Variation as a Statistical Hypothesis Test
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/1504.06507