Tree Structure-Aware Few-Shot Image Classification via Hierarchical Aggregation

TL;DR

This paper introduces a hierarchical tree structure-aware method that adaptively aggregates features from pretext tasks to significantly improve few-shot image classification performance across multiple benchmarks.

Contribution

The proposed HTS method uniquely learns task relationships and adaptively combines pretext task features to enhance few-shot learning, achieving state-of-the-art results.

Findings

HTS significantly improves few-shot classification accuracy.

The method achieves new state-of-the-art on four benchmark datasets.

Adaptive feature aggregation enhances transferability to novel classes.

Abstract

In this paper, we mainly focus on the problem of how to learn additional feature representations for few-shot image classification through pretext tasks (e.g., rotation or color permutation and so on). This additional knowledge generated by pretext tasks can further improve the performance of few-shot learning (FSL) as it differs from human-annotated supervision (i.e., class labels of FSL tasks). To solve this problem, we present a plug-in Hierarchical Tree Structure-aware (HTS) method, which not only learns the relationship of FSL and pretext tasks, but more importantly, can adaptively select and aggregate feature representations generated by pretext tasks to maximize the performance of FSL tasks. A hierarchical tree constructing component and a gated selection aggregating component is introduced to construct the tree structure and find richer transferable knowledge that can rapidly adapt to novel classes with a few labeled images. Extensive experiments show that our HTS can significantly enhance multiple few-shot methods to achieve new state-of-the-art performance on four benchmark datasets. The code is available at: https://github.com/remiMZ/HTS-ECCV22.