Mutual Exclusive Modulator for Long-Tailed Recognition

TL;DR

This paper introduces a mutual exclusive modulator that classifies long-tailed data into three groups to improve recognition performance, addressing class imbalance by estimating group probabilities and guiding classifiers.

Contribution

The paper proposes a novel mutual exclusive modulator that predicts class group probabilities to enhance long-tailed recognition, a new approach focusing on data volume clues.

Findings

Achieves competitive results on ImageNet-LT, Place-LT, and iNaturalist 2018 datasets.

Effectively reduces class imbalance impact by separating categories into three groups.

Outperforms several existing methods in long-tailed recognition benchmarks.

Abstract

The long-tailed recognition (LTR) is the task of learning high-performance classifiers given extremely imbalanced training samples between categories. Most of the existing works address the problem by either enhancing the features of tail classes or re-balancing the classifiers to reduce the inductive bias. In this paper, we try to look into the root cause of the LTR task, i.e., training samples for each class are greatly imbalanced, and propose a straightforward solution. We split the categories into three groups, i.e., many, medium and few, according to the number of training images. The three groups of categories are separately predicted to reduce the difficulty for classification. This idea naturally arises a new problem of how to assign a given sample to the right class groups? We introduce a mutual exclusive modulator which can estimate the probability of an image belonging to each group. Particularly, the modulator consists of a light-weight module and learned with a mutual exclusive objective. Hence, the output probabilities of the modulator encode the data volume clues of the training dataset. They are further utilized as prior information to guide the prediction of the classifier. We conduct extensive experiments on multiple datasets, e.g., ImageNet-LT, Place-LT and iNaturalist 2018 to evaluate the proposed approach. Our method achieves competitive performance compared to the state-of-the-art benchmarks.