[人工智能]CVPR2022 底层视觉 | 图像处理

图像去雾

Self-Augmented Unpaired Image Dehazing via Density and Depth Decomposition [pdf][code]

Towards Multi-Domain Single Image Dehazing via Test-Time Training [pdf]

Image Dehazing Transformer With Transmission-Aware 3D Position Embedding [pdf] [supp][arXiv] [code]

图像去雨

Unpaired Deep Image Deraining Using Dual Contrastive Learning [pdf] [arXiv]

Dreaming To Prune Image Deraining Networks [pdf] [supp]

Unsupervised Deraining: Where Contrastive Learning Meets Self-Similarity [pdf] [supp] [arXiv]

超分辨率

BasicVSR++: Improving Video Super-Resolution With Enhanced Propagation and Alignment [pdf] [supp]

Investigating Tradeoffs in Real-World Video Super-Resolution [pdf] [supp] [arXiv]

Texture-Based Error Analysis for Image Super-Resolution [pdf]

Task Decoupled Framework for Reference-Based Super-Resolution [pdf] [supp]

Reflash Dropout in Image Super-Resolution [pdf] [supp] [arXiv]

Learning the Degradation Distribution for Blind Image Super-Resolution [pdf] [arXiv]

Details or Artifacts: A Locally Discriminative Learning Approach to Realistic Image Super-Resolution [pdf] [supp] [arXiv][code]

Dual Adversarial Adaptation for Cross-Device Real-World Image Super-Resolution [pdf] [arXiv]

VideoINR: Learning Video Implicit Neural Representation for Continuous Space-Time Super-Resolution [pdf] [supp]

Discrete Cosine Transform Network for Guided Depth Map Super-Resolution [pdf] [supp] [arXiv]

Blind Image Super-Resolution With Elaborate Degradation Modeling on Noise and Kernel [pdf] [supp] [arXiv][code]

A Text Attention Network for Spatial Deformation Robust Scene Text Image Super-Resolution [pdf] [supp] [arXiv][code]

Deep Constrained Least Squares for Blind Image Super-Resolution [pdf] [supp] [arXiv]

Look Back and Forth: Video Super-Resolution With Explicit Temporal Difference Modeling [pdf] [arXiv]

Towards Bidirectional Arbitrary Image Rescaling: Joint Optimization and Cycle Idempotence [paper]

HyperTransformer: A Textural and Spectral Feature Fusion Transformer for Pansharpening [paper][code]

HDNet: High-resolution Dual-domain Learning for Spectral Compressive Imaging [paper]

低光照增强

Toward Fast, Flexible, and Robust Low-Light Image Enhancement [paper] [code]

Deep Color Consistent Network for Low-Light Image Enhancement [paper]

URetinex-Net: Retinex-based Deep Unfolding Network for Low-light Image Enhancement [paper][code]

图像去噪

AP-BSN: Self-Supervised Denoising for Real-World Images via Asymmetric PD and Blind-Spot Network [paper] [code]

IDR: Self-Supervised Image Denoising via Iterative Data Refinement [paper] [code]

Blind2Unblind: Self-Supervised Image Denoising with Visible Blind Spots [paper] [code]

图像复原

Exploring and Evaluating Image Restoration Potential in Dynamic Scenes [paper]

Come-Closer-Diffuse-Faster: Accelerating Conditional Diffusion Models for Inverse Problems through Stochastic Contraction [paper]

Mask-guided Spectral-wise Transformer for Efficient Hyperspectral Image Reconstruction [paper] [code]

Restormer: Efficient Transformer for High-Resolution Image Restoration [paper] [code]

Event-based Video Reconstruction via Potential-assisted Spiking Neural Network [paper]

文章来源：CVPR2022 底层视觉 | 图像处理 - 知乎