Fast Adaptive Meta-Learning for Few-Shot Image Generation
- 5 May 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 24 (15209210) , 2205-2217
Abstract
Generative Adversarial Networks (GANs) are capable of effectively synthesising new realistic images and estimating the potential distribution of samples utilising adversarial learning. Nevertheless, conventional GANs require a large amount of training data samples to produce plausible results. Inspired by the capacity for humans to quickly learn new concepts from a small number of examples, several meta-learning approaches for the few-shot datasets are presented. However, most of meta-learning algorithms are designed to tackle few-shot classification and reinforcement learning tasks. Moreover, the existing meta-learning models for image generation are complex, thereby affecting the length of training time required. Fast Adaptive Meta-Learning (FAML) based on GAN and the encoder network is proposed in this study for few-shot image generation. This model demonstrates the capability to generate new realistic images from previously unseen target classes with only a small number of examples required. With 10 times faster convergence, FAML requires only one-fourth of the trainable parameters in comparison baseline models by training a simpler network with conditional feature vectors from the encoder, while increasing the number of generator iterations. The visualisation results are demonstrated in the paper. This model is able to improve few-shot image generation with the lowest FID score, highest IS, and comparable LPIPS to MNIST, Omniglot, VGG-Faces, and mini ImageNet datasets. The source code is available on https://github.com/phaphuang/FAML .Keywords
Funding Information
- National Key Research and Development Program of China (2018YFC0807105)
- Science and Technology Committee of Shanghai Municipality (17DZ1101003, 18511106602, 18DZ2252300)
- Open Funding Project of the State Key Laboratory of Bioreactor Engineering
- East China University of Science and Technology, Shanghai, China
- International College of Digital Innovation
- Chiang Mai University, Thailand
This publication has 41 references indexed in Scilit:
- Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks2017
- Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network2017
- Generalized Deep Transfer Networks for Knowledge Propagation in Heterogeneous DomainsACM Transactions on Multimedia Computing, Communications, and Applications, 2016
- Joint Intermodal and Intramodal Label Transfers for Extremely Rare or Unseen ClassesIEEE Transactions on Pattern Analysis and Machine Intelligence, 2016
- Rethinking the Inception Architecture for Computer Vision2016
- Deep Learning Face Attributes in the Wild2015
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- Weakly-Shared Deep Transfer Networks for Heterogeneous-Domain Knowledge Propagation2015
- Metalearning: a survey of trends and technologiesArtificial Intelligence Review, 2013
- One-shot learning of object categoriesIEEE Transactions on Pattern Analysis and Machine Intelligence, 2006