MIS-Boost: Multiple Instance Selection Boosting

TL;DR

MIS-Boost introduces a flexible multiple instance learning method that learns discriminative prototypes through explicit selection, outperforming existing methods and effectively identifying meaningful image regions in large-scale classification tasks.

Contribution

It proposes a novel MIL approach that learns prototypes in a continuous space with data-driven control over prototype quantity, enhancing flexibility and performance.

Findings

Outperforms state-of-the-art MIL methods on benchmark datasets.

Automatically identifies visually meaningful image regions.

Demonstrates effectiveness in large-scale image classification.

Abstract

In this paper, we present a new multiple instance learning (MIL) method, called MIS-Boost, which learns discriminative instance prototypes by explicit instance selection in a boosting framework. Unlike previous instance selection based MIL methods, we do not restrict the prototypes to a discrete set of training instances but allow them to take arbitrary values in the instance feature space. We also do not restrict the total number of prototypes and the number of selected-instances per bag; these quantities are completely data-driven. We show that MIS-Boost outperforms state-of-the-art MIL methods on a number of benchmark datasets. We also apply MIS-Boost to large-scale image classification, where we show that the automatically selected prototypes map to visually meaningful image regions.