Uncertainty in Minimum Cost Multicuts for Image and Motion Segmentation

TL;DR

This paper introduces a probabilistic approach to quantify uncertainties in minimum cost multicuts, enhancing image and motion segmentation tasks by providing decision confidence measures and evaluating their impact on segmentation quality.

Contribution

It presents a novel probabilistic formulation for uncertainty estimation in minimum cost multicuts, improving the interpretability and reliability of segmentation results.

Findings

Uncertainty measures correlate with segmentation quality.

Sparsification based on uncertainty improves segmentation accuracy.

Method tested successfully on multiple benchmark datasets.

Abstract

The minimum cost lifted multicut approach has proven practically good performance in a wide range of applications such as image decomposition, mesh segmentation, multiple object tracking, and motion segmentation. It addresses such problems in a graph-based model, where real-valued costs are assigned to the edges between entities such that the minimum cut decomposes the graph into an optimal number of segments. Driven by a probabilistic formulation of minimum cost multicuts, we provide a measure for the uncertainties of the decisions made during the optimization. We argue that access to such uncertainties is crucial for many practical applications and conduct an evaluation by means of sparsifications on three different, widely used datasets in the context of image decomposition (BSDS-500) and motion segmentation (DAVIS2016 and FBMS59) in terms of variation of information (VI) and Rand index (RI).