Reinforcing Generated Images via Meta-learning for One-Shot Fine-Grained Visual Recognition

TL;DR

This paper introduces a meta-learning approach that combines GAN-generated images with original images to enhance one-shot fine-grained visual recognition, resulting in improved accuracy and increased diversity of training data.

Contribution

It proposes a novel meta-learning framework with a MetaIRNet to reinforce generated images, improving one-shot fine-grained recognition performance.

Findings

Consistent improvement over baselines on benchmark datasets

Reinforced images exhibit greater diversity

MetaIRNet effectively combines generated and original images

Abstract

One-shot fine-grained visual recognition often suffers from the problem of having few training examples for new fine-grained classes. To alleviate this problem, off-the-shelf image generation techniques based on Generative Adversarial Networks (GANs) can potentially create additional training images. However, these GAN-generated images are often not helpful for actually improving the accuracy of one-shot fine-grained recognition. In this paper, we propose a meta-learning framework to combine generated images with original images, so that the resulting "hybrid" training images improve one-shot learning. Specifically, the generic image generator is updated by a few training instances of novel classes, and a Meta Image Reinforcing Network (MetaIRNet) is proposed to conduct one-shot fine-grained recognition as well as image reinforcement. Our experiments demonstrate consistent improvement over baselines on one-shot fine-grained image classification benchmarks. Furthermore, our analysis shows that the reinforced images have more diversity compared to the original and GAN-generated images.