Test-Time Amendment with a Coarse Classifier for Fine-Grained Classification

TL;DR

This paper introduces Hierarchical Ensembles (HiE), a post-hoc method leveraging label hierarchy to enhance fine-grained classification accuracy and reduce mistake severity, applicable at test-time and in semi-supervised settings.

Contribution

The paper proposes a novel, simple post-hoc correction method using label hierarchy to improve fine-grained classification performance and mistake severity reduction.

Findings

Achieves state-of-the-art results on iNaturalist-19 and tieredImageNet-H datasets.

Significantly reduces mistake severity while improving top-1 accuracy.

Effective in semi-supervised settings with less training data.

Abstract

We investigate the problem of reducing mistake severity for fine-grained classification. Fine-grained classification can be challenging, mainly due to the requirement of domain expertise for accurate annotation. However, humans are particularly adept at performing coarse classification as it requires relatively low levels of expertise. To this end, we present a novel approach for Post-Hoc Correction called Hierarchical Ensembles (HiE) that utilizes label hierarchy to improve the performance of fine-grained classification at test-time using the coarse-grained predictions. By only requiring the parents of leaf nodes, our method significantly reduces avg. mistake severity while improving top-1 accuracy on the iNaturalist-19 and tieredImageNet-H datasets, achieving a new state-of-the-art on both benchmarks. We also investigate the efficacy of our approach in the semi-supervised setting. Our approach brings notable gains in top-1 accuracy while significantly decreasing the severity of mistakes as training data decreases for the fine-grained classes. The simplicity and post-hoc nature of HiE renders it practical to be used with any off-the-shelf trained model to improve its predictions further.