Domain-Invariant Projection Learning for Zero-Shot Recognition

TL;DR

This paper introduces a domain-invariant projection learning model for zero-shot recognition that improves the robustness of classifying unseen objects by aligning feature and semantic spaces through a novel self-reconstruction task and shared semantic structures.

Contribution

The paper proposes a new ZSL model with a domain-invariant feature self-reconstruction task and superclass formation in semantic space, along with a novel iterative optimization algorithm.

Findings

Outperforms state-of-the-art ZSL methods significantly.

Effective domain alignment via shared semantic structures.

Robust projection learning with a new min-min optimization formulation.

Abstract

Zero-shot learning (ZSL) aims to recognize unseen object classes without any training samples, which can be regarded as a form of transfer learning from seen classes to unseen ones. This is made possible by learning a projection between a feature space and a semantic space (e.g. attribute space). Key to ZSL is thus to learn a projection function that is robust against the often large domain gap between the seen and unseen classes. In this paper, we propose a novel ZSL model termed domain-invariant projection learning (DIPL). Our model has two novel components: (1) A domain-invariant feature self-reconstruction task is introduced to the seen/unseen class data, resulting in a simple linear formulation that casts ZSL into a min-min optimization problem. Solving the problem is non-trivial, and a novel iterative algorithm is formulated as the solver, with rigorous theoretic algorithm analysis provided. (2) To further align the two domains via the learned projection, shared semantic structure among seen and unseen classes is explored via forming superclasses in the semantic space. Extensive experiments show that our model outperforms the state-of-the-art alternatives by significant margins.