Home

Awesome

Awesome-ECCV2022-Low-Level-Vision

A Collection of Papers and Codes in ECCV2022 related to Low-Level Vision

Related collections for low-level vision

Catalogue

<!-- - [Image Rescaling](#Rescaling) --> <!-- - [Demosaicing](#Demosaicing) --> <!-- - [Image Stitching](#Stitching) -->

<a name="ImageRetoration"></a>

Image Restoration - 图像恢复

Simple Baselines for Image Restoration

D2HNet: Joint Denoising and Deblurring with Hierarchical Network for Robust Night Image Restoration

Seeing Far in the Dark with Patterned Flash

BayesCap: Bayesian Identity Cap for Calibrated Uncertainty in Frozen Neural Networks

Improving Image Restoration by Revisiting Global Information Aggregation

Fast Two-step Blind Optical Aberration Correction

VQFR: Blind Face Restoration with Vector-Quantized Dictionary and Parallel Decoder

RAWtoBit: A Fully End-to-end Camera ISP Network

Transform your Smartphone into a DSLR Camera: Learning the ISP in the Wild

Single Frame Atmospheric Turbulence Mitigation: A Benchmark Study and A New Physics-Inspired Transformer Model

Modeling Mask Uncertainty in Hyperspectral Image Reconstruction

TAPE: Task-Agnostic Prior Embedding for Image Restoration

DRCNet: Dynamic Image Restoration Contrastive Network

ART-SS: An Adaptive Rejection Technique for Semi-Supervised Restoration for Adverse Weather-Affected Images

Spectrum-Aware and Transferable Architecture Search for Hyperspectral Image Restoration

Seeing through a Black Box: Toward High-Quality Terahertz Imaging via Subspace-and-Attention Guided Restoration

JPEG Artifacts Removal via Contrastive Representation Learning

Zero-Shot Learning for Reflection Removal of Single 360-Degree Image

Overexposure Mask Fusion: Generalizable Reverse ISP Multi-Step Refinement

<a name="VideoRestoration"></a>

Video Restoration

Video Restoration Framework and Its Meta-Adaptations to Data-Poor Conditions

<a name="SuperResolution"></a>

Super Resolution - 超分辨率

<a name="ImageSuperResolution"></a>

Image Super Resolution

ARM: Any-Time Super-Resolution Method

Dynamic Dual Trainable Bounds for Ultra-low Precision Super-Resolution Networks

CADyQ : Contents-Aware Dynamic Quantization for Image Super Resolution

Image Super-Resolution with Deep Dictionary

Perception-Distortion Balanced ADMM Optimization for Single-Image Super-Resolution

Adaptive Patch Exiting for Scalable Single Image Super-Resolution

Learning Series-Parallel Lookup Tables for Efficient Image Super-Resolution

MuLUT: Cooperating Mulitple Look-Up Tables for Efficient Image Super-Resolution

Efficient Long-Range Attention Network for Image Super-resolution

Compiler-Aware Neural Architecture Search for On-Mobile Real-time Super-Resolution

Restore Globally, Refine Locally: A Mask-Guided Scheme to Accelerate Super-Resolution Networks

Learning Mutual Modulation for Self-Supervised Cross-Modal Super-Resolution

Self-Supervised Learning for Real-World Super-Resolution from Dual Zoomed Observations

Efficient and Degradation-Adaptive Network for Real-World Image Super-Resolution

D2C-SR: A Divergence to Convergence Approach for Real-World Image Super-Resolution

MM-RealSR: Metric Learning based Interactive Modulation for Real-World Super-Resolution

KXNet: A Model-Driven Deep Neural Network for Blind Super-Resolution

From Face to Natural Image: Learning Real Degradation for Blind Image Super-Resolution

Unfolded Deep Kernel Estimation for Blind Image Super-Resolution

Uncertainty Learning in Kernel Estimation for Multi-stage Blind Image Super-Resolution

Super-Resolution by Predicting Offsets: An Ultra-Efficient Super-Resolution Network for Rasterized Images

Reference-based Image Super-Resolution with Deformable Attention Transformer

RRSR:Reciprocal Reference-Based Image Super-Resolution with Progressive Feature Alignment and Selection

Boosting Event Stream Super-Resolution with a Recurrent Neural Network

HST: Hierarchical Swin Transformer for Compressed Image Super-resolution

Swin2SR: SwinV2 Transformer for Compressed Image Super-Resolution and Restoration

Fast Nearest Convolution for Real-Time Efficient Image Super-Resolution

<a name="VideoSuperResolution"></a>

Video Super Resolution

Learning Spatiotemporal Frequency-Transformer for Compressed Video Super-Resolution

A Codec Information Assisted Framework for Efficient Compressed Video Super-Resolution

Real-RawVSR: Real-World Raw Video Super-Resolution with a Benchmark Dataset

<a name="Denoising"></a>

Denoising - 去噪

<a name="ImageDenoising"></a>

Image Denoising

Deep Semantic Statistics Matching (D2SM) Denoising Network

Fast and High Quality Image Denoising via Malleable Convolution

<a name="VideoDenoising"></a>

Video Denoising

Unidirectional Video Denoising by Mimicking Backward Recurrent Modules with Look-ahead Forward Ones

TempFormer: Temporally Consistent Transformer for Video Denoising

<a name="Deblurring"></a>

Deblurring - 去模糊

<a name="ImageDeblurring"></a>

Image Deblurring

Learning Degradation Representations for Image Deblurring

Stripformer: Strip Transformer for Fast Image Deblurring

Animation from Blur: Multi-modal Blur Decomposition with Motion Guidance

United Defocus Blur Detection and Deblurring via Adversarial Promoting Learning

Realistic Blur Synthesis for Learning Image Deblurring

Event-based Fusion for Motion Deblurring with Cross-modal Attention

Event-Guided Deblurring of Unknown Exposure Time Videos

<a name="VideoDeblurring"></a>

Video Deblurring

Spatio-Temporal Deformable Attention Network for Video Deblurring

Efficient Video Deblurring Guided by Motion Magnitude

ERDN: Equivalent Receptive Field Deformable Network for Video Deblurring

DeMFI: Deep Joint Deblurring and Multi-Frame Interpolation with Flow-Guided Attentive Correlation and Recursive Boosting

Towards Real-World Video Deblurring by Exploring Blur Formation Process

<a name="Decomposition"></a>

Image Decomposition

Blind Image Decomposition

<a name="Deraining"></a>

Deraining - 去雨

Not Just Streaks: Towards Ground Truth for Single Image Deraining

Rethinking Video Rain Streak Removal: A New Synthesis Model and a Deraining Network with Video Rain Prior

<a name="Dehazing"></a>

Dehazing - 去雾

Frequency and Spatial Dual Guidance for Image Dehazing

Perceiving and Modeling Density for Image Dehazing

Boosting Supervised Dehazing Methods via Bi-Level Patch Reweighting

Unpaired Deep Image Dehazing Using Contrastive Disentanglement Learning

<a name="Demoireing"></a>

Demoireing - 去摩尔纹

Towards Efficient and Scale-Robust Ultra-High-Definition Image Demoireing

<a name="HDR"></a>

HDR Imaging / Multi-Exposure Image Fusion - HDR图像生成 / 多曝光图像融合

Exposure-Aware Dynamic Weighted Learning for Single-Shot HDR Imaging

Ghost-free High Dynamic Range Imaging with Context-aware Transformer

Selective TransHDR: Transformer-Based Selective HDR Imaging Using Ghost Region Mask

HDR-Plenoxels: Self-Calibrating High Dynamic Range Radiance Fields

Towards Real-World HDRTV Reconstruction: A Data Synthesis-Based Approach

<a name="Fusion"></a>

Image Fusion

FusionVAE: A Deep Hierarchical Variational Autoencoder for RGB Image Fusion

Recurrent Correction Network for Fast and Efficient Multi-modality Image Fusion

Neural Image Representations for Multi-Image Fusion and Layer Separation

Fusion from Decomposition: A Self-Supervised Decomposition Approach for Image Fusion

<a name="FrameInterpolation"></a>

Frame Interpolation - 插帧

Real-Time Intermediate Flow Estimation for Video Frame Interpolation

FILM: Frame Interpolation for Large Motion

Video Interpolation by Event-driven Anisotropic Adjustment of Optical Flow

Learning Cross-Video Neural Representations for High-Quality Frame Interpolation

Deep Bayesian Video Frame Interpolation

A Perceptual Quality Metric for Video Frame Interpolation

DeMFI: Deep Joint Deblurring and Multi-Frame Interpolation with Flow-Guided Attentive Correlation and Recursive Boosting

<a name="STVSR"></a>

Spatial-Temporal Video Super-Resolution

Towards Interpretable Video Super-Resolution via Alternating Optimization

<a name="Enhancement"></a>

Image Enhancement - 图像增强

Local Color Distributions Prior for Image Enhancement

SepLUT: Separable Image-adaptive Lookup Tables for Real-time Image Enhancement

Neural Color Operators for Sequential Image Retouching

Deep Fourier-Based Exposure Correction Network with Spatial-Frequency Interaction

Uncertainty Inspired Underwater Image Enhancement

NEST: Neural Event Stack for Event-Based Image Enhancement

<a name="LowLight"></a>

Low-Light Image Enhancement

LEDNet: Joint Low-light Enhancement and Deblurring in the Dark

Unsupervised Night Image Enhancement: When Layer Decomposition Meets Light-Effects Suppression

<a name="Harmonization"></a>

Image Harmonization - 图像协调

Harmonizer: Learning to Perform White-Box Image and Video Harmonization

DCCF: Deep Comprehensible Color Filter Learning Framework for High-Resolution Image Harmonization

Semantic-Guided Multi-Mask Image Harmonization

Spatial-Separated Curve Rendering Network for Efficient and High-Resolution Image Harmonization

<a name="Inpainting"></a>

Image Completion/Inpainting - 图像修复

Learning Prior Feature and Attention Enhanced Image Inpainting

Perceptual Artifacts Localization for Inpainting

High-Fidelity Image Inpainting with GAN Inversion

Unbiased Multi-Modality Guidance for Image Inpainting

Image Inpainting with Cascaded Modulation GAN and Object-Aware Training

Inpainting at Modern Camera Resolution by Guided PatchMatch with Auto-Curation

Diverse Image Inpainting with Normalizing Flow

Hourglass Attention Network for Image Inpainting

Perceptual Artifacts Localization for Inpainting

Don't Forget Me: Accurate Background Recovery for Text Removal via Modeling Local-Global Context

The Surprisingly Straightforward Scene Text Removal Method with Gated Attention and Region of Interest Generation: A Comprehensive Prominent Model Analysis

Video Inpainting

Error Compensation Framework for Flow-Guided Video Inpainting

Flow-Guided Transformer for Video Inpainting

<a name="Colorization"></a>

Image Colorization - 图像上色

Eliminating Gradient Conflict in Reference-based Line-art Colorization

Bridging the Domain Gap towards Generalization in Automatic Colorization

CT2: Colorization Transformer via Color Tokens

PalGAN: Image Colorization with Palette Generative Adversarial Networks

BigColor: Colorization using a Generative Color Prior for Natural Images

Semantic-Sparse Colorization Network for Deep Exemplar-Based Colorization

ColorFormer: Image Colorization via Color Memory Assisted Hybrid-Attention Transformer

L-CoDer: Language-Based Colorization with Color-Object Decoupling Transformer

Colorization for In Situ Marine Plankton Images

<a name="Matting"></a>

Image Matting - 图像抠图

TransMatting: Enhancing Transparent Objects Matting with Transformers

One-Trimap Video Matting

<a name="ShadowRemoval"></a>

Shadow Removal - 阴影消除

Style-Guided Shadow Removal

<a name="ImageCompression"></a>

Image Compression - 图像压缩

Optimizing Image Compression via Joint Learning with Denoising

Implicit Neural Representations for Image Compression

Expanded Adaptive Scaling Normalization for End to End Image Compression

Content-Oriented Learned Image Compression

Contextformer: A Transformer with Spatio-Channel Attention for Context Modeling in Learned Image Compression

Content Adaptive Latents and Decoder for Neural Image Compression

Video Compression

AlphaVC: High-Performance and Efficient Learned Video Compression

CANF-VC: Conditional Augmented Normalizing Flows for Video Compression

Neural Video Compression Using GANs for Detail Synthesis and Propagation

<a name="ImageQualityAssessment"></a>

Image Quality Assessment - 图像质量评价

FAST-VQA: Efficient End-to-end Video Quality Assessment with Fragment Sampling

Shift-tolerant Perceptual Similarity Metric

Telepresence Video Quality Assessment

A Perceptual Quality Metric for Video Frame Interpolation

<a name="Relighting"></a>

Relighting/Delighting

Deep Portrait Delighting

Geometry-Aware Single-Image Full-Body Human Relighting

NeRF for Outdoor Scene Relighting

Physically-Based Editing of Indoor Scene Lighting from a Single Image

<a name="StyleTransfer"></a>

Style Transfer - 风格迁移

CCPL: Contrastive Coherence Preserving Loss for Versatile Style Transfer

Image-Based CLIP-Guided Essence Transfer

Learning Graph Neural Networks for Image Style Transfer

WISE: Whitebox Image Stylization by Example-based Learning

Language-Driven Artistic Style Transfer

MoDA: Map Style Transfer for Self-Supervised Domain Adaptation of Embodied Agents

JoJoGAN: One Shot Face Stylization

EleGANt: Exquisite and Locally Editable GAN for Makeup Transfer

RamGAN: Region Attentive Morphing GAN for Region-Level Makeup Transfer

<a name="ImageEditing"></a>

Image Editing - 图像编辑

Context-Consistent Semantic Image Editing with Style-Preserved Modulation

GAN with Multivariate Disentangling for Controllable Hair Editing

Paint2Pix: Interactive Painting based Progressive Image Synthesis and Editing

High-fidelity GAN Inversion with Padding Space

Text2LIVE: Text-Driven Layered Image and Video Editing

IntereStyle: Encoding an Interest Region for Robust StyleGAN Inversion

Style Your Hair: Latent Optimization for Pose-Invariant Hairstyle Transfer via Local-Style-Aware Hair Alignment

HairNet: Hairstyle Transfer with Pose Changes

End-to-End Visual Editing with a Generatively Pre-trained Artist

The Anatomy of Video Editing: A Dataset and Benchmark Suite for AI-Assisted Video Editing

Scraping Textures from Natural Images for Synthesis and Editing

VQGAN-CLIP: Open Domain Image Generation and Editing with Natural Language Guidance

Editing Out-of-Domain GAN Inversion via Differential Activations

ChunkyGAN: Real Image Inversion via Segments

FairStyle: Debiasing StyleGAN2 with Style Channel Manipulations

A Style-Based GAN Encoder for High Fidelity Reconstruction of Images and Videos

Rayleigh EigenDirections (REDs): Nonlinear GAN latent space traversals for multidimensional features

<a name=ImageGeneration></a>

Image Generation/Synthesis / Image-to-Image Translation - 图像生成/合成/转换

Text-to-Image / Text Guided / Multi-Modal

TIPS: Text-Induced Pose Synthesis

TISE: A Toolbox for Text-to-Image Synthesis Evaluation

Learning Visual Styles from Audio-Visual Associations

Multimodal Conditional Image Synthesis with Product-of-Experts GANs

NÜWA: Visual Synthesis Pre-training for Neural visUal World creAtion

Make-a-Scene: Scene-Based Text-to-Image Generation with Human Priors

Trace Controlled Text to Image Generation

Audio-Driven Stylized Gesture Generation with Flow-Based Model

No Token Left Behind: Explainability-Aided Image Classification and Generation

Image-to-Image / Image Guided

End-to-end Graph-constrained Vectorized Floorplan Generation with Panoptic Refinement

ManiFest: Manifold Deformation for Few-shot Image Translation

VecGAN: Image-to-Image Translation with Interpretable Latent Directions

DynaST: Dynamic Sparse Transformer for Exemplar-Guided Image Generation

Cross Attention Based Style Distribution for Controllable Person Image Synthesis

Vector Quantized Image-to-Image Translation

URUST: Ultra-high-resolution unpaired stain transformation via Kernelized Instance Normalization

General Object Pose Transformation Network from Unpaired Data

Unpaired Image Translation via Vector Symbolic Architectures

Supervised Attribute Information Removal and Reconstruction for Image Manipulation

Bi-Level Feature Alignment for Versatile Image Translation and Manipulation

Multi-Curve Translator for High-Resolution Photorealistic Image Translation

CoGS: Controllable Generation and Search from Sketch and Style

AgeTransGAN for Facial Age Transformation with Rectified Performance Metrics

Others for image generation

StyleLight: HDR Panorama Generation for Lighting Estimation and Editing

Accelerating Score-based Generative Models with Preconditioned Diffusion Sampling

GAN Cocktail: mixing GANs without dataset access

Compositional Visual Generation with Composable Diffusion Models

Adaptive-Feature-Interpolation-for-Low-Shot-Image-Generation

StyleHEAT: One-Shot High-Resolution Editable Talking Face Generation via Pretrained StyleGAN

WaveGAN: An Frequency-aware GAN for High-Fidelity Few-shot Image Generation

FakeCLR: Exploring Contrastive Learning for Solving Latent Discontinuity in Data-Efficient GANs

Auto-regressive Image Synthesis with Integrated Quantization

PixelFolder: An Efficient Progressive Pixel Synthesis Network for Image Generation

DeltaGAN: Towards Diverse Few-shot Image Generation with Sample-Specific Delta

Generator Knows What Discriminator Should Learn in Unconditional GANs

Hierarchical Semantic Regularization of Latent Spaces in StyleGANs

FurryGAN: High Quality Foreground-aware Image Synthesis

Improving GANs for Long-Tailed Data through Group Spectral Regularization

Exploring Gradient-based Multi-directional Controls in GANs

Improved Masked Image Generation with Token-Critic

Weakly-Supervised Stitching Network for Real-World Panoramic Image Generation

Any-Resolution Training for High-Resolution Image Synthesis

BIPS: Bi-modal Indoor Panorama Synthesis via Residual Depth-Aided Adversarial Learning

Few-Shot Image Generation with Mixup-Based Distance Learning

StyleGAN-Human: A Data-Centric Odyssey of Human Generation

StyleFace: Towards Identity-Disentangled Face Generation on Megapixels

Contrastive Learning for Diverse Disentangled Foreground Generation

BLT: Bidirectional Layout Transformer for Controllable Layout Generation

Entropy-Driven Sampling and Training Scheme for Conditional Diffusion Generation

Unleashing Transformers: Parallel Token Prediction with Discrete Absorbing Diffusion for Fast High-Resolution Image Generation from Vector-Quantized Codes

DuelGAN: A Duel between Two Discriminators Stabilizes the GAN Training

<a name="VideoGeneration"></a>

Video Generation

Long Video Generation with Time-Agnostic VQGAN and Time-Sensitive Transformer

Controllable Video Generation through Global and Local Motion Dynamics

Fast-Vid2Vid: Spatial-Temporal Compression for Video-to-Video Synthesis

Synthesizing Light Field Video from Monocular Video

StoryDALL-E: Adapting Pretrained Text-to-Image Transformers for Story Continuation

Motion Transformer for Unsupervised Image Animation

Sound-Guided Semantic Video Generation

Layered Controllable Video Generation

Diverse Generation from a Single Video Made Possible

Semantic-Aware Implicit Neural Audio-Driven Video Portrait Generation

EAGAN: Efficient Two-Stage Evolutionary Architecture Search for GANs

BlobGAN: Spatially Disentangled Scene Representations

<a name="Others"></a>

Others

Learning Local Implicit Fourier Representation for Image Warping

Dress Code: High-Resolution Multi-Category Virtual Try-On

High-Resolution Virtual Try-On with Misalignment and Occlusion-Handled Conditions

Single Stage Virtual Try-on via Deformable Attention Flows

Outpainting by Queries

Watermark Vaccine: Adversarial Attacks to Prevent Watermark Removal

Efficient Meta-Tuning for Content-aware Neural Video Delivery

Human-centric Image Cropping with Partition-aware and Content-preserving Features

CelebV-HQ: A Large-Scale Video Facial Attributes Dataset

Learning Dynamic Facial Radiance Fields for Few-Shot Talking Head Synthesis

Responsive Listening Head Generation: A Benchmark Dataset and Baseline

Contrastive Monotonic Pixel-Level Modulation

AutoTransition: Learning to Recommend Video Transition Effects

Bringing Rolling Shutter Images Alive with Dual Reversed Distortion

Learning Object Placement via Dual-path Graph Completion

DeepMCBM: A Deep Moving-camera Background Model

Mind the Gap in Distilling StyleGANs

StyleSwap: Style-Based Generator Empowers Robust Face Swapping

Geometric Representation Learning for Document Image Rectification

Studying Bias in GANs through the Lens of Race

On the Robustness of Quality Measures for GANs

TREND: Truncated Generalized Normal Density Estimation of Inception Embeddings for GAN Evaluation

Hallucinating Pose-Compatible Scenes

<!-- **** - Paper: - Code: -->