Dual Expert Distillation Network for Generalized Zero-Shot Learning

TL;DR

This paper introduces the Dual Expert Distillation Network (DEDN) for generalized zero-shot learning, leveraging dual experts for coarse and fine attribute modeling and a dual attention backbone to improve visual-semantic understanding, achieving state-of-the-art results.

Contribution

The paper proposes a novel dual expert framework and a dual attention backbone to better model attributes and utilize visual semantic knowledge in zero-shot learning.

Findings

Achieves state-of-the-art performance on benchmark datasets.

Effectively models asymmetric attributes and channel information.

Demonstrates the benefits of cooperative distillation between experts.

Abstract

Zero-shot learning has consistently yielded remarkable progress via modeling nuanced one-to-one visual-attribute correlation. Existing studies resort to refining a uniform mapping function to align and correlate the sample regions and subattributes, ignoring two crucial issues: 1) the inherent asymmetry of attributes; and 2) the unutilized channel information. This paper addresses these issues by introducing a simple yet effective approach, dubbed Dual Expert Distillation Network (DEDN), where two experts are dedicated to coarse- and fine-grained visual-attribute modeling, respectively. Concretely, one coarse expert, namely cExp, has a complete perceptual scope to coordinate visual-attribute similarity metrics across dimensions, and moreover, another fine expert, namely fExp, consists of multiple specialized subnetworks, each corresponds to an exclusive set of attributes. Two experts cooperatively distill from each other to reach a mutual agreement during training. Meanwhile, we further equip DEDN with a newly designed backbone network, i.e., Dual Attention Network (DAN), which incorporates both region and channel attention information to fully exploit and leverage visual semantic knowledge. Experiments on various benchmark datasets indicate a new state-of-the-art.