Disentangling Hardness from Noise: An Uncertainty-Driven Model-Agnostic Framework for Long-Tailed Remote Sensing Classification

TL;DR

This paper introduces DUAL, a model-agnostic framework that disentangles uncertainty types to improve long-tailed remote sensing classification by effectively handling hard samples and noise.

Contribution

It proposes a novel uncertainty-driven approach that separates epistemic and aleatoric uncertainties to enhance class imbalance handling in remote sensing tasks.

Findings

DUAL outperforms strong baselines like TGN and SADE.

The framework effectively distinguishes hard samples from noisy data.

Ablation studies validate the importance of uncertainty components.

Abstract

Long-Tailed distributions are pervasive in remote sensing due to the inherently imbalanced occurrence of grounded objects. However, a critical challenge remains largely overlooked, i.e., disentangling hard tail data samples from noisy ambiguous ones. Conventional methods often indiscriminately emphasize all low-confidence samples, leading to overfitting on noisy data. To bridge this gap, building upon Evidential Deep Learning, we propose a model-agnostic uncertainty-aware framework termed DUAL, which dynamically disentangles prediction uncertainty into Epistemic Uncertainty (EU) and Aleatoric Uncertainty (AU). Specifically, we introduce EU as an indicator of sample scarcity to guide a reweighting strategy for hard-to-learn tail samples, while leveraging AU to quantify data ambiguity, employing an adaptive label smoothing mechanism to suppress the impact of noise. Extensive experiments on multiple datasets across various backbones demonstrate the effectiveness and generalization of our framework, surpassing strong baselines such as TGN and SADE. Ablation studies provide further insights into the crucial choices of our design.