Zero-Shot Learning from Adversarial Feature Residual to Compact Visual Feature

TL;DR

This paper introduces an adversarial network that synthesizes compact visual features for zero-shot learning, reducing feature overlap and improving recognition accuracy across multiple datasets.

Contribution

A novel adversarial framework with residual generation and feature selection for more discriminative and less overlapped unseen-class features in zero-shot learning.

Findings

Achieves 1.2-13.2% higher accuracy than state-of-the-art methods.

Reduces overlap of unseen-class feature distributions.

Demonstrates effectiveness on six benchmark datasets.

Abstract

Recently, many zero-shot learning (ZSL) methods focused on learning discriminative object features in an embedding feature space, however, the distributions of the unseen-class features learned by these methods are prone to be partly overlapped, resulting in inaccurate object recognition. Addressing this problem, we propose a novel adversarial network to synthesize compact semantic visual features for ZSL, consisting of a residual generator, a prototype predictor, and a discriminator. The residual generator is to generate the visual feature residual, which is integrated with a visual prototype predicted via the prototype predictor for synthesizing the visual feature. The discriminator is to distinguish the synthetic visual features from the real ones extracted from an existing categorization CNN. Since the generated residuals are generally numerically much smaller than the distances among all the prototypes, the distributions of the unseen-class features synthesized by the proposed network are less overlapped. In addition, considering that the visual features from categorization CNNs are generally inconsistent with their semantic features, a simple feature selection strategy is introduced for extracting more compact semantic visual features. Extensive experimental results on six benchmark datasets demonstrate that our method could achieve a significantly better performance than existing state-of-the-art methods by 1.2-13.2% in most cases.