Reinforcing Generated Images via Meta-learning for One-Shot Fine-Grained Visual Recognition
arXiv:2204.10689 · doi:10.1109/TPAMI.2022.3167112
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.
arXiv commentAccepted to PAMI 2022. arXiv admin note: substantial text overlap with arXiv:1911.07164
References in corpus (12)
- Adam: A Method for Stochastic Optimization
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- Prototypical Networks for Few-shot Learning
- mixup: Beyond Empirical Risk Minimization
- Spectral Normalization for Generative Adversarial Networks
- A Learned Representation For Artistic Style
- Wasserstein GAN
- Modulating early visual processing by language
- Delta-encoder: an effective sample synthesis method for few-shot object recognition
- Swapping Autoencoder for Deep Image Manipulation
- Freeze the Discriminator: a Simple Baseline for Fine-Tuning GANs