Integrated Generalized Zero-Shot Learning for Fine-Grained Classification

TL;DR

This paper introduces an integrated approach combining embedding learning and feature synthesizing with local and global features, using a novel attention mechanism and mutual learning to improve fine-grained generalized zero-shot learning performance.

Contribution

It presents a unified network with a dense attention mechanism and mutual learning, effectively leveraging local and global features for fine-grained GZSL.

Findings

Outperforms existing methods on benchmark datasets

Effective use of attribute-guided local features

Reduces bias towards source domain during testing

Abstract

Embedding learning (EL) and feature synthesizing (FS) are two of the popular categories of fine-grained GZSL methods. EL or FS using global features cannot discriminate fine details in the absence of local features. On the other hand, EL or FS methods exploiting local features either neglect direct attribute guidance or global information. Consequently, neither method performs well. In this paper, we propose to explore global and direct attribute-supervised local visual features for both EL and FS categories in an integrated manner for fine-grained GZSL. The proposed integrated network has an EL sub-network and a FS sub-network. Consequently, the proposed integrated network can be tested in two ways. We propose a novel two-step dense attention mechanism to discover attribute-guided local visual features. We introduce new mutual learning between the sub-networks to exploit mutually beneficial information for optimization. Moreover, we propose to compute source-target class similarity based on mutual information and transfer-learn the target classes to reduce bias towards the source domain during testing. We demonstrate that our proposed method outperforms contemporary methods on benchmark datasets.