Twofold Debiasing Enhances Fine-Grained Learning with Coarse Labels

TL;DR

This paper introduces a Twofold Debiasing method that enhances fine-grained recognition in coarse-to-fine learning by improving feature extraction and distribution calibration, leading to state-of-the-art results.

Contribution

The paper proposes a novel Twofold Debiasing approach with feature fusion and distribution calibration to improve fine-grained learning from coarse labels.

Findings

Achieves state-of-the-art performance on five benchmark datasets.

Effectively enhances fine-grained feature extraction.

Reduces overfitting caused by biased distributions.

Abstract

The Coarse-to-Fine Few-Shot (C2FS) task is designed to train models using only coarse labels, then leverages a limited number of subclass samples to achieve fine-grained recognition capabilities. This task presents two main challenges: coarse-grained supervised pre-training suppresses the extraction of critical fine-grained features for subcategory discrimination, and models suffer from overfitting due to biased distributions caused by limited fine-grained samples. In this paper, we propose the Twofold Debiasing (TFB) method, which addresses these challenges through detailed feature enhancement and distribution calibration. Specifically, we introduce a multi-layer feature fusion reconstruction module and an intermediate layer feature alignment module to combat the model's tendency to focus on simple predictive features directly related to coarse-grained supervision, while neglecting complex fine-grained level details. Furthermore, we mitigate the biased distributions learned by the fine-grained classifier using readily available coarse-grained sample embeddings enriched with fine-grained information. Extensive experiments conducted on five benchmark datasets demonstrate the efficacy of our approach, achieving state-of-the-art results that surpass competitive methods.