Augmented Cyclic Consistency Regularization for Unpaired Image-to-Image Translation
arXiv:2003.00187
Abstract
Unpaired image-to-image (I2I) translation has received considerable attention in pattern recognition and computer vision because of recent advancements in generative adversarial networks (GANs). However, due to the lack of explicit supervision, unpaired I2I models often fail to generate realistic images, especially in challenging datasets with different backgrounds and poses. Hence, stabilization is indispensable for GANs and applications of I2I translation. Herein, we propose Augmented Cyclic Consistency Regularization (ACCR), a novel regularization method for unpaired I2I translation. Our main idea is to enforce consistency regularization originating from semi-supervised learning on the discriminators leveraging real, fake, reconstructed, and augmented samples. We regularize the discriminators to output similar predictions when fed pairs of original and perturbed images. We qualitatively clarify why consistency regularization on fake and reconstructed samples works well. Quantitatively, our method outperforms the consistency regularized GAN (CR-GAN) in real-world translations and demonstrates efficacy against several data augmentation variants and cycle-consistent constraints.
arXiv commentAccepted to ICPR2020
References in corpus (11)
- Spectral Normalization for Generative Adversarial Networks
- Instance Normalization: The Missing Ingredient for Fast Stylization
- Improved Regularization of Convolutional Neural Networks with Cutout
- Toward Multimodal Image-to-Image Translation
- A Learned Representation For Artistic Style
- Random Erasing Data Augmentation
- Learning to Discover Cross-Domain Relations with Generative Adversarial Networks
- U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
- Augmented CycleGAN: Learning Many-to-Many Mappings from Unpaired Data
- Learning Discrete Representations via Information Maximizing Self-Augmented Training
- On Convergence and Stability of GANs