HiFaceGAN: Face Renovation via Collaborative Suppression and Replenishment (2020-05-11T00:00:00.000000Z)

TL;DR

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, is presented.

Abstract

Existing face restoration researches typically rely on either the image degradation prior or explicit guidance labels for training, which often lead to limited generalization ability over real-world images with heterogeneous degradation and rich background contents. In this paper, we investigate a 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 formulate 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 our HiFaceGAN over a wide range of challenging restoration subtasks, demonstrating its versatility, robustness and generalization ability towards real-world face processing applications. Code is available at https://github.com/Lotayou/Face-Renovation.

Authors

Siwei Ma

3 papers

Wen Gao

4 papers

Peiran Ren

4 papers

TL;DR

Abstract

Authors

References67 items

Enhanced Blind Face Restoration With Multi-Exemplar Images and Adaptive Spatial Feature Fusion

Blind Image Restoration without Prior Knowledge

Efficient and Interpretable Deep Blind Image Deblurring Via Algorithm Unrolling

Analyzing and Improving the Image Quality of StyleGAN

SEAN: Image Synthesis With Semantic Region-Adaptive Normalization

A Comprehensive Benchmark for Single Image Compression Artifact Reduction

Variational Denoising Network: Toward Blind Noise Modeling and Removal

Progressive Face Super-Resolution via Attention to Facial Landmark

LIP: Local Importance-Based Pooling

DeblurGAN-v2: Deblurring (Orders-of-Magnitude) Faster and Better

MaskGAN: Towards Diverse and Interactive Facial Image Manipulation

Blind Deblurring using Deep Learning: A Survey

Deep Learning Face Hallucination via Attributes Transfer and Enhancement

Real Image Denoising With Feature Attention

Pixel-Adaptive Convolutional Neural Networks

Feedback Network for Image Super-Resolution

Semantic Image Synthesis With Spatially-Adaptive Normalization

Face Hallucination via Coarse-to-Fine Recursive Kernel Regression Structure

Sequential Gating Ensemble Network for Noise Robust Multiscale Face Restoration

A Style-Based Generator Architecture for Generative Adversarial Networks

Face Hallucination Techniques: A Survey

Face Super-Resolution Guided by Facial Component Heatmaps

ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks

Video-to-Video Synthesis

Deep Learning for Single Image Super-Resolution: A Brief Review

GANimation: Anatomically-aware Facial Animation from a Single Image

Enhanced Image Decoding via Edge-Preserving Generative Adversarial Networks

Face Hallucination Using Cascaded Super-Resolution and Identity Priors

Multi-Scale Face Restoration With Sequential Gating Ensemble Network

Learning Warped Guidance for Blind Face Restoration

Graph-Based Blind Image Deblurring From a Single Photograph

The Unreasonable Effectiveness of Deep Features as a Perceptual Metric

Super-FAN: Integrated Facial Landmark Localization and Super-Resolution of Real-World Low Resolution Faces in Arbitrary Poses with GANs

High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs

FSRNet: End-to-End Learning Face Super-Resolution with Facial Priors

DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks

Toward Multimodal Image-to-Image Translation

Progressive Growing of GANs for Improved Quality, Stability, and Variation

Wavelet-SRNet: A Wavelet-Based CNN for Multi-scale Face Super Resolution

Category-Specific Object Image Denoising

Enhanced Deep Residual Networks for Single Image Super-Resolution

GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium

When Image Denoising Meets High-Level Vision Tasks: A Deep Learning Approach

Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks

Arbitrary Style Transfer in Real-Time with Adaptive Instance Normalization

Image Restoration Techniques: A Survey

Image-to-Image Translation with Conditional Adversarial Networks

Least Squares Generative Adversarial Networks

Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network

Perceptual Losses for Real-Time Style Transfer and Super-Resolution

Rethinking the Inception Architecture for Computer Vision

Accurate Image Super-Resolution Using Very Deep Convolutional Networks

Compression Artifacts Reduction by a Deep Convolutional Network

Image Super-Resolution Using Deep Convolutional Networks

Very Deep Convolutional Networks for Large-Scale Image Recognition

One millisecond face alignment with an ensemble of regression trees

Generative Adversarial Nets

Making a “Completely Blind” Image Quality Analyzer

ImageNet classification with deep convolutional neural networks

Close the loop: Joint blind image restoration and recognition with sparse representation prior

Image quality assessment: from error visibility to structural similarity

Multiscale structural similarity for image quality assessment

Component Semantic Prior Guided Generative Adversarial Network for Face Super-Resolution

Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks

Blind Image Restoration

Vision: A computational investigation in the human representation of visual information

Learned-Miller, and

Field of Study

Journal Information

Name

Venue Information

Name

Type

URL

Alternate Names