Enhancing Few-Shot Image Classification through Learnable Multi-Scale Embedding and Attention Mechanisms

TL;DR

This paper introduces a multi-scale embedding and attention-based approach to improve few-shot image classification, capturing diverse features and refining representations for better accuracy across multiple datasets.

Contribution

It proposes a novel multi-output embedding network with attention mechanisms and learnable stage weights, enhancing feature diversity and robustness in few-shot learning.

Findings

Achieved high accuracy on MiniImageNet and FC100 datasets.

Improved performance in cross-domain few-shot tasks.

Outperformed state-of-the-art methods in experiments.

Abstract

In the context of few-shot classification, the goal is to train a classifier using a limited number of samples while maintaining satisfactory performance. However, traditional metric-based methods exhibit certain limitations in achieving this objective. These methods typically rely on a single distance value between the query feature and support feature, thereby overlooking the contribution of shallow features. To overcome this challenge, we propose a novel approach in this paper. Our approach involves utilizing a multi-output embedding network that maps samples into distinct feature spaces. The proposed method extracts feature vectors at different stages, enabling the model to capture both global and abstract features. By utilizing these diverse feature spaces, our model enhances its performance. Moreover, employing a self-attention mechanism improves the refinement of features at each stage, leading to even more robust representations and improved overall performance. Furthermore, assigning learnable weights to each stage significantly improved performance and results. We conducted comprehensive evaluations on the MiniImageNet and FC100 datasets, specifically in the 5-way 1-shot and 5-way 5-shot scenarios. Additionally, we performed cross-domain tasks across eight benchmark datasets, achieving high accuracy in the testing domains. These evaluations demonstrate the efficacy of our proposed method in comparison to state-of-the-art approaches. https://github.com/FatemehAskari/MSENet