HiFaceGAN: Face Renovation via Collaborative Suppression and Replenishment

Abstract

Existing face restoration researches typically relies on either the degradation prior or explicit guidance labels for training, which often results in limited generalization ability over real-world images with heterogeneous degradations and rich background contents. In this paper, we investigate the more challenging and practical "dual-blind" version of the problem by lifting the requirements on both types of prior, termed as "Face Renovation"(FR). Specifically, we formulated FR as a semantic-guided generation problem and tackle it with a collaborative suppression and replenishment (CSR) approach. This leads to HiFaceGAN, a multi-stage framework containing several nested CSR units that progressively replenish facial details based on the hierarchical semantic guidance extracted from the front-end content-adaptive suppression modules. Extensive experiments on both synthetic and real face images have verified the superior performance of HiFaceGAN over a wide range of challenging restoration subtasks, demonstrating its versatility, robustness and generalization ability towards real-world face processing applications.

Published in ACM Multimedia 2020

References in corpus (14)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks
• Deep Learning for Single Image Super-Resolution: A Brief Review
• Toward Multimodal Image-to-Image Translation
• Variational Denoising Network: Toward Blind Noise Modeling and Removal
• Graph-Based Blind Image Deblurring From a Single Photograph
• Video-to-Video Synthesis
• Progressive Face Super-Resolution via Attention to Facial Landmark
• A Comprehensive Benchmark for Single Image Compression Artifacts Reduction
• MaskGAN: Towards Diverse and Interactive Facial Image Manipulation
• Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization
• Blind Deblurring using Deep Learning: A Survey
• Learning Warped Guidance for Blind Face Restoration
• Blind Image Restoration without Prior Knowledge

Cited by in corpus (17)

• Semi-Cycled Generative Adversarial Networks for Real-World Face Super-Resolution
• Low-field magnetic resonance image enhancement via stochastic image quality transfer
• Towards Fine-grained Human Pose Transfer with Detail Replenishing Network
• LTT-GAN: Looking Through Turbulence by Inverting GANs
• Towards Real-World Blind Face Restoration with Generative Facial Prior
• GAN Prior Embedded Network for Blind Face Restoration in the Wild
• Super-Resolving Face Image by Facial Parsing Information
• IFQA: Interpretable Face Quality Assessment
• Simultaneous Face Hallucination and Translation for Thermal to Visible Face Verification using Axial-GAN
• Progressive Semantic-Aware Style Transformation for Blind Face Restoration
• OSDFace: One-Step Diffusion Model for Face Restoration
• Deep Learning-based Face Super-Resolution: A Survey
• Universal Face Restoration With Memorized Modulation
• Analysis and Benchmarking of Extending Blind Face Image Restoration to Videos
• Network Architecture Search for Face Enhancement
• Implicit Subspace Prior Learning for Dual-Blind Face Restoration
• Generative Iris Prior Embedded Transformer for Iris Restoration