Fine-Grained Visual Categorization using Meta-Learning Optimization with Sample Selection of Auxiliary Data

TL;DR

This paper introduces MetaFGNet, a meta-learning based approach for fine-grained visual categorization that optimizes model pre-training for better adaptation to target tasks and includes a sample selection scheme from auxiliary data.

Contribution

The paper proposes MetaFGNet with a novel meta-learning objective and a sample selection scheme to improve FGVC performance using auxiliary data.

Findings

MetaFGNet outperforms existing methods on benchmark datasets.

Sample selection from auxiliary data enhances fine-tuning effectiveness.

Meta-learning objective improves adaptation to target FGVC tasks.

Abstract

Fine-grained visual categorization (FGVC) is challenging due in part to the fact that it is often difficult to acquire an enough number of training samples. To employ large models for FGVC without suffering from overfitting, existing methods usually adopt a strategy of pre-training the models using a rich set of auxiliary data, followed by fine-tuning on the target FGVC task. However, the objective of pre-training does not take the target task into account, and consequently such obtained models are suboptimal for fine-tuning. To address this issue, we propose in this paper a new deep FGVC model termed MetaFGNet. Training of MetaFGNet is based on a novel regularized meta-learning objective, which aims to guide the learning of network parameters so that they are optimal for adapting to the target FGVC task. Based on MetaFGNet, we also propose a simple yet effective scheme for selecting more useful samples from the auxiliary data. Experiments on benchmark FGVC datasets show the efficacy of our proposed method.