CVPR 2024 超分辨率（super-resolution）方向上接收论文总结

CVPR 2024

CVPR 2024 会议将于 2024 年 6 月 17 日至 21 日在美国西雅图举行。

今年共提交了11532份有效论文，2719篇论文被接收，录用率为23.6%。2719篇被接收论文中，具体录用情况：

• Poster论文有2305篇，占比84.8%。CVPR组委会将以Poster形式展示论文。

• Poster（Highlight）被所在地区主席认为特别有趣或具有创新性的论文，今年共有324篇（11.9%）。

• Oral论文共有90篇，占比3.3%。录用者可以15分钟介绍自己的论文。

现将超分辨率方向上接收的论文汇总如下，遗漏之处还请大家斧正。

图像超分

1. Self-Adaptive Reality-Guided Diffusion for Artifact-Free Super-Resolution
   • Paper: https://arxiv.org/abs/2403.16643
   • Code: https://github.com/ProAirVerse/Self-Adaptive-Guidance-Diffusion
   • Keywords: Diffusion; Artifact-Free
2. Text-guided Explorable Image Super-resolution
   • Paper: https://arxiv.org/abs/2403.01124
   • Code: https://github.com/KVGandikota/Text-guidedSR
   • Keywords: Zero-shot; Diffusion; Text-to-image T2I
3. Training Generative Image Super-Resolution Models by Wavelet-Domain Losses Enables Better Control of Artifacts
   • Paper: https://arxiv.org/abs/2402.19215
   • Code: https://github.com/mandalinadagi/WGSR
   • Keywords: Wavelet-Domain Losses
4. CFAT: Unleashing TriangularWindows for Image Super-resolution
   • Paper: https://arxiv.org/abs/2403.16143
   • Code: https://github.com/rayabhisek123/CFAT
   • Keywords: Transformer
5. AdaBM: On-the-Fly Adaptive Bit Mapping for Image Super-Resolution
   • Paper: https://arxiv.org/abs/2404.03296
   • Code: https://github.com/Cheeun/AdaBM
   • Keywords: Quantization 量化
6. Transcending the Limit of Local Window: Advanced Super-Resolution Transformer with Adaptive Token Dictionary
   • Paper: https://arxiv.org/abs/2401.08209
   • Code: https://github.com/LabShuHangGU/Adaptive-Token-Dictionary
   • Keywords: Transformer
7. SinSR: Diffusion-Based Image Super-Resolution in a Single Step
   • Paper: https://arxiv.org/abs/2311.14760
   • Code: https://github.com/wyf0912/SinSR
   • Keywords: Diffusion
8. Image Processing GNN: Breaking Rigidity in Super-Resolution
   • Paper:
   • Code:
   • Keywords: GNN
9. Low-Res Leads the Way: Improving Generalization for Super-Resolution by Self-Supervised Learning
   • Paper: https://arxiv.org/abs/2403.02601
   • Code: https://haoyuchen.com/LWay
   • Keywords:
   • Blog: CVPR 2024 | 涨点！LWay：通过自监督学习提高超分辨率的泛化能力
   • Team: Zhensong Zhang （张镇嵩）Huawei Noah’s Ark Lab
10. Boosting Flow-based Generative Super-Resolution Models via Learned Prior
    • Paper: https://arxiv.org/abs/2403.10988
    • Code: https://github.com/liyuantsao/FlowSR-LP
    • Keywords: Flow
11. Learning Coupled Dictionaries from Unpaired Data for Image Super-Resolution
    • Paper: https://junchenglee.com/paper/CVPR.pdf
    • Code:
    • Keywords: Autoencoder
12. Navigating Beyond Dropout: An Intriguing Solution towards Generalizable Image Super-Resolution
    • Paper: https://arxiv.org/abs/2402.18929
    • Code: https://github.com/Dreamzz5/Simple-Align
    • Keywords: Dropout
    • Blog: CVPR 2024 | 超越Dropout：通用图像超分辨率的有趣解决方案
13. CoDi: Conditional Diffusion Distillation for Higher-Fidelity and Faster Image Generation
    • Paper: https://arxiv.org/abs/2310.01407
    • Code: https://fast-codi.github.io/
    • HuggingFace: https://huggingface.co/spaces/MKFMIKU/CoDi
    • Keywords: Multi-task, Diffusion, Distillation
    • Blog:https://zhuanlan.zhihu.com/p/686266981

盲超分 / 真实世界 / 参考

1. A Dynamic Kernel Prior Model for Unsupervised Blind Image Super-Resolution
   • Paper: https://arxiv.org/abs/2404.15620
   • Code: https://github.com/XYLGroup/DKP
   • Keywords: Unsupervised, Blind
2. CDFormer: When Degradation Prediction Embraces Diffusion Model for Blind Image Super-Resolution
   • Paper:
   • Code:
   • Keywords: Diffusion; Blind
3. Diffusion-based Blind Text Image Super-Resolution
   • Paper: https://arxiv.org/abs/2312.08886
   • Code:
   • Keywords: Blind; Diffusion; Real-world; Text
4. SeD: Semantic-Aware Discriminator for Image Super-Resolution
   • Paper: https://arxiv.org/abs/2402.19387
   • Code: https://github.com/lbc12345/SeD
   • Keywords: Real SR; pretrained vision models (PVMs);
5. Uncertainty-Aware Source-Free Adaptive Image Super-Resolution with Wavelet Augmentation Transformer
   • Paper: https://arxiv.org/abs/2303.17783
   • Code:
   • Keywords: Unsupervised Domain Adaptation (UDA); Transformer; Real-world
6. Universal Robustness via Median Random Smoothing for Real-World Super-Resolution
   • Paper:
   • Code:
   • Keywords: Real-World
7. SeeSR: Towards Semantics-Aware Real-World Image Super-Resolution
   • Paper: https://arxiv.org/abs/2311.16518
   • Code: https://github.com/cswry/SeeSR
   • Keywords: Real-World; Prompt; text-to-image (T2I)
   • Blog: CVPR 2024 真实超分 SeeSR: Towards Semantics-Aware Real-World Image Super-Resolution
   • Team: Lei Zhang
8. CoSeR: Bridging Image and Language for Cognitive Super-Resolution
   • Paper: https://arxiv.org/abs/2311.16512
   • Code: https://github.com/VINHYU/CoSeR
   • Keywords: Text-to-image;
9. Building Bridges across Spatial and Temporal Resolutions: Reference-Based Super-Resolution via Change Priors and Conditional Diffusion Model
   • Paper: https://arxiv.org/abs/2403.17460
   • Code: https://github.com/dongrunmin/RefDiff
   • Keywords: Diffusion, RefSR

视频超分

1. FMA-Net: Flow Guided Dynamic Filtering and Iterative Feature Refinement with Multi-Attention for Joint Video Super-Resolution and Deblurring (Oral)
   • Paper: http://arxiv.org/abs/2401.03707
   • Code: https://kaist-viclab.github.io/fmanet-site/
   • Keywords: SR & Deblurring
2. Learning Spatial Adaptation and Temporal Coherence in Diffusion Models for Video Super-Resolution
   • Paper: https://arxiv.org/abs/2403.17000
   • Code:
   • Keywords: Diffusion
3. Enhancing Video Super-Resolution via Implicit Resampling-based Alignment (Highlight)
   • Paper: https://arxiv.org/abs/2305.00163
   • Code: https://github.com/kai422/IART
   • Keywords: Video
4. Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolution
   • Paper: https://arxiv.org/abs/2312.06640
   • Code: https://github.com/sczhou/Upscale-A-Video
   • Keywords: Diffusion
5. Video Super-Resolution Transformer with Masked Inter&Intra-Frame Attention
   • Paper: https://arxiv.org/abs/2401.06312
   • Code: https://github.com/LabShuHangGU/MIA-VSR
   • Keywords: Transformer

数据集

1. Continuous Optical Zooming: A Benchmark for Arbitrary-Scale Image Super-Resolution in Real World
   • Paper:
   • Code:
   • Keywords: Real World; Arbitrary-Scale

特殊场景

1. Super-Resolution Reconstruction from Bayer-Pattern Spike Streams
   • Paper:
   • Code: https://github.com/csycdong/CSCSR
   • Keywords: Bayer-Pattern Spike Streams
   • Team: 张健 北京大学 VILLA 实验室;
2. Rethinking Diffusion Model for Multi-Contrast MRI Super-Resolution
   • Paper: https://arxiv.org/abs/2404.04785
   • Code: https://github.com/GuangYuanKK/DiffMSR
   • Keywords: MRI; Diffusion
3. Bilateral Event Mining and Complementary for Event Stream Super-Resolution
   • Paper:
   • Code: https://github.com/Lqm26/BMCNet-ESR
   • Keywords: Event Stream
4. APISR: Anime Production Inspired Real-World Anime Super-Resolution
   • Paper: https://arxiv.org/abs/2403.01598
   • Code: https://github.com/Kiteretsu77/APISR
   • Keywords: Anime 动漫超分
5. CycleINR: Cycle Implicit Neural Representation for Arbitrary-Scale Volumetric Super-Resolution of Medical Data
   • Paper: https://arxiv.org/abs/2404.04878
   • Code:
   • Keywords: Arbitrary-Scale; Medical
6. Neural Super-Resolution for Real-time Rendering with Radiance Demodulation
   • Paper: https://arxiv.org/abs/2308.06699
   • Code: https://github.com/Riga2/NSRD
   • Keywords: Real-time; Video;
7. PFStorer: Personalized Face Restoration and Super-Resolution
   • Paper: https://arxiv.org/abs/2403.08436
   • Code:
   • Keywords: Face
8. Learning Large-Factor EM Image Super-Resolution with Generative Priors
   • Paper:
   • Code: https://github.com/jtshou/GPEMSR
   • Keywords: Electron microscopy (EM)

总结

从本届接收的论文来看，Diffusion model 和 Text 引入语义文本信息是大的热点，单纯的超分基本已经绝迹，一般必须带上特殊场景或背景。

参考资料

1. CVPR 2024 Accepted Papers
2. Image/Video Super-Resolution(图像超分辨率)
3. CVPR2024 Super-Resolution
4. CVPR2024 底层视觉 | 图像处理