Hybrid-Attention based Decoupled Metric Learning for Zero-Shot Image Retrieval

TL;DR

This paper introduces a decoupled metric learning framework with attention mechanisms to improve zero-shot image retrieval by enhancing discrimination and generalization, outperforming existing methods on benchmarks.

Contribution

It proposes a novel decoupled metric learning approach with attention modules to address zero-shot challenges, emphasizing discrimination and generalization.

Findings

Outperforms state-of-the-art methods on benchmark datasets.

Effectively enhances visual discrimination in zero-shot retrieval.

Demonstrates the importance of addressing zero-shot specific problems.

Abstract

In zero-shot image retrieval (ZSIR) task, embedding learning becomes more attractive, however, many methods follow the traditional metric learning idea and omit the problems behind zero-shot settings. In this paper, we first emphasize the importance of learning visual discriminative metric and preventing the partial/selective learning behavior of learner in ZSIR, and then propose the Decoupled Metric Learning (DeML) framework to achieve these individually. Instead of coarsely optimizing an unified metric, we decouple it into multiple attention-specific parts so as to recurrently induce the discrimination and explicitly enhance the generalization. And they are mainly achieved by our object-attention module based on random walk graph propagation and the channel-attention module based on the adversary constraint, respectively. We demonstrate the necessity of addressing the vital problems in ZSIR on the popular benchmarks, outperforming the state-of-theart methods by a significant margin. Code is available at http://www.bhchen.cn