f-VAEGAN-D2: A Feature Generating Framework for Any-Shot Learning

TL;DR

This paper introduces f-VAEGAN-D2, a unified feature generating framework that improves zero-shot and few-shot learning by leveraging unlabeled data and combining VAE and GAN techniques for highly discriminative and interpretable CNN features.

Contribution

It proposes a novel conditional generative model that integrates VAE and GANs, utilizing unlabeled data to enhance feature generation in any-shot learning scenarios.

Findings

Achieves state-of-the-art results on five datasets including ImageNet.

Learns highly discriminative CNN features for zero- and few-shot learning.

Features are interpretable through visualization and textual explanations.

Abstract

When labeled training data is scarce, a promising data augmentation approach is to generate visual features of unknown classes using their attributes. To learn the class conditional distribution of CNN features, these models rely on pairs of image features and class attributes. Hence, they can not make use of the abundance of unlabeled data samples. In this paper, we tackle any-shot learning problems i.e. zero-shot and few-shot, in a unified feature generating framework that operates in both inductive and transductive learning settings. We develop a conditional generative model that combines the strength of VAE and GANs and in addition, via an unconditional discriminator, learns the marginal feature distribution of unlabeled images. We empirically show that our model learns highly discriminative CNN features for five datasets, i.e. CUB, SUN, AWA and ImageNet, and establish a new state-of-the-art in any-shot learning, i.e. inductive and transductive (generalized) zero- and few-shot learning settings. We also demonstrate that our learned features are interpretable: we visualize them by inverting them back to the pixel space and we explain them by generating textual arguments of why they are associated with a certain label.