Synthesized Classifiers for Zero-Shot Learning

TL;DR

This paper introduces a novel zero-shot learning approach that synthesizes classifiers by aligning semantic and model spaces using phantom classes, leading to improved recognition of unseen object classes.

Contribution

It proposes a manifold learning framework with phantom classes to synthesize classifiers, enhancing zero-shot learning accuracy on large-scale datasets.

Findings

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

Achieves high accuracy on the large-scale ImageNet Fall 2011 dataset.

Demonstrates the effectiveness of manifold alignment in zero-shot recognition.

Abstract

Given semantic descriptions of object classes, zero-shot learning aims to accurately recognize objects of the unseen classes, from which no examples are available at the training stage, by associating them to the seen classes, from which labeled examples are provided. We propose to tackle this problem from the perspective of manifold learning. Our main idea is to align the semantic space that is derived from external information to the model space that concerns itself with recognizing visual features. To this end, we introduce a set of "phantom" object classes whose coordinates live in both the semantic space and the model space. Serving as bases in a dictionary, they can be optimized from labeled data such that the synthesized real object classifiers achieve optimal discriminative performance. We demonstrate superior accuracy of our approach over the state of the art on four benchmark datasets for zero-shot learning, including the full ImageNet Fall 2011 dataset with more than 20,000 unseen classes.