On Considering Uncertainty and Alternatives in Low-Level Vision

TL;DR

This paper explores how Bayesian methods can model uncertainty in image segmentation, allowing for a probabilistic representation of multiple segmentation alternatives to improve understanding and decision-making in low-level vision tasks.

Contribution

It introduces a Bayesian framework for representing and managing uncertainty in image segmentation, including detailed methods for constructing probability distributions over segmentation alternatives.

Findings

Bayesian formalism effectively models segmentation uncertainty.

Statistical image models can approximate distributions over segmentation spaces.

The approach enhances understanding of uncertainty levels in segmentation tasks.

Abstract

In this paper we address the uncertainty issues involved in the low-level vision task of image segmentation. Researchers in computer vision have worked extensively on this problem, in which the goal is to partition (or segment) an image into regions that are homogeneous or uniform in some sense. This segmentation is often utilized by some higher level process, such as an object recognition system. We show that by considering uncertainty in a Bayesian formalism, we can use statistical image models to build an approximate representation of a probability distribution over a space of alternative segmentations. We give detailed descriptions of the various levels of uncertainty associated with this problem, discuss the interaction of prior and posterior distributions, and provide the operations for constructing this representation.