Focal Frequency Loss for Image Reconstruction and Synthesis

Abstract

Image reconstruction and synthesis have witnessed remarkable progress thanks to the development of generative models. Nonetheless, gaps could still exist between the real and generated images, especially in the frequency domain. In this study, we show that narrowing gaps in the frequency domain can ameliorate image reconstruction and synthesis quality further. We propose a novel focal frequency loss, which allows a model to adaptively focus on frequency components that are hard to synthesize by down-weighting the easy ones. This objective function is complementary to existing spatial losses, offering great impedance against the loss of important frequency information due to the inherent bias of neural networks. We demonstrate the versatility and effectiveness of focal frequency loss to improve popular models, such as VAE, pix2pix, and SPADE, in both perceptual quality and quantitative performance. We further show its potential on StyleGAN2.

arXiv commentICCV 2021. GitHub: https://github.com/EndlessSora/focal-frequency-loss Project page: https://www.mmlab-ntu.com/project/ffl/index.html

References in corpus (20)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Auto-Encoding Variational Bayes
• Conditional Generative Adversarial Nets
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
• Large Scale GAN Training for High Fidelity Natural Image Synthesis
• Spectral Normalization for Generative Adversarial Networks
• Instance Normalization: The Missing Ingredient for Fast Stylization
• Self-Attention Generative Adversarial Networks
• Conditional Image Generation with PixelCNN Decoders
• Semi-Supervised Learning with Deep Generative Models
• Which Training Methods for GANs do actually Converge?
• Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks
• On the Spectral Bias of Neural Networks
• Learning to Predict Layout-to-image Conditional Convolutions for Semantic Image Synthesis
• SWAGAN: A Style-based Wavelet-driven Generative Model
• Unsupervised Real-world Image Super Resolution via Domain-distance Aware Training
• DeeperForensics Challenge 2020 on Real-World Face Forgery Detection: Methods and Results
• RDA: Robust Domain Adaptation via Fourier Adversarial Attacking