Full search page →

SeeSR: degradation-aware semantic prompts (CVPR 2024)

Optional · paper guide 10 of 11 for DLCV 09A.1 · Blind and Real-World Image Super-Resolution. Citation numbers refer to the reference list at the bottom of this page.

Navigate: this page and this track: Navigate: this page and this track

On this page

Supplementary notes · 3

  1. The high-order degradation pipeline, step by step
  2. Sinc filters and ringing
  3. Why the kernel is hard to identify

Paper guides · 11

  1. SRMD: one network for many known degradations (CVPR 2018)
  2. IKC: iterative kernel correction (CVPR 2019)
  3. KernelGAN: kernel estimation with an internal GAN (NeurIPS 2019)
  4. DAN: unfolding the alternating optimization (NeurIPS 2020)
  5. DASR: contrastive degradation representations (CVPR 2021)
  6. BSRGAN: a practical degradation model (ICCV 2021)
  7. Real-ESRGAN: high-order degradation with pure synthetic data (ICCVW 2021)
  8. StableSR: a time-aware encoder for a frozen diffusion prior (IJCV 2024)
  9. DiffBIR: remove the degradation, then regenerate (ECCV 2024)
  10. SeeSR: degradation-aware semantic prompts (CVPR 2024)
  11. SUPIR: scaling up the prior, the data and the text (CVPR 2024)

Deep Learning for Computer Vision

  1. 02 · Deep Image Classification
View the whole course tree →

Challenge

A blurred object can be consistent with several semantic interpretations. A text-to-image prior guided by the wrong label may generate sharp, coherent detail for the wrong object. SeeSR treats semantic conditioning as a degradation-sensitive estimation problem [29]. A general captioner or tagger trained on clean photographs is not automatically reliable on blurry, noisy LR inputs.

The design combines image structure, discrete text tags and continuous semantic features. The three conditions have different jobs: structure constrains position, tags identify concepts, and feature embeddings retain information that a short list of labels discards. The method also changes the starting latent so that pure-noise sampling does not encourage unnecessary texture in smooth input regions.

Method

Train the degradation-aware prompt extractor separately, then use semantic prompts and low-resolution image conditioning to steer the diffusion restorer.
Original paper figure 2 — Train the degradation-aware prompt extractor separately, then use semantic prompts and low-resolution image conditioning to steer the diffusion restorer.

Figure source: [29], arXiv:2311.16518v2, PDF page 5. Original PDF.

Degradation-aware prompt extractor (DAPE). A pretrained image-tagging model [78] supplies a frozen clean-image teacher and an adapted degraded-image branch. Given an HR image xx and synthetic LR partner yy, their representation and tag-logit vectors are aligned through

LDAPE=∥fyrep−fxrep∥22+λ∥fylogit−fxlogit∥22.\mathcal L_{\mathrm{DAPE}}=\|f^{\mathrm{rep}}_y-f^{\mathrm{rep}}_x\|_2^2+\lambda\|f^{\mathrm{logit}}_y-f^{\mathrm{logit}}_x\|_2^2.

Here frepf^{\mathrm{rep}} denotes an image representation, flogitf^{\mathrm{logit}} the vector of pre-threshold tag scores, and λ\lambda the relative weight. Each pair of vectors has matching dimensionality. The clean branch is fixed; the degraded branch learns to produce clean-like semantics despite damage. Low-rank adapters are used to limit adaptation cost [79]. This supervision is available because the training degradation supplies aligned HR–LR partners.

Hard and soft prompts. Thresholded predicted tags become text that enters the frozen text encoder. The unthresholded image representation becomes a soft condition. Increasing the tag threshold trades recall for precision; the soft representation partly avoids the information discarded by that threshold. The two are complementary, but neither can infer an unrecognizable object with certainty.

Image and representation control. A ControlNet-style path receives image features [42]. Representation cross-attention modules follow text cross-attention and inject the soft semantic features into the denoiser. In attention notation, queries come from denoising features, while keys and values come from the semantic condition. This gives spatially evolving image features access to a richer representation than a tag string alone. The pretrained diffusion backbone stays frozen; the added image encoder, control path and attention modules learn with the usual noise-prediction objective.

LR embedding at inference. Let zyz_y be the autoencoded, enlarged LR image and αˉT\bar\alpha_T the terminal signal coefficient of the pretrained noise schedule. Start from

zT=αˉT zy+1−αˉT ϵ,ϵ∼N(0,I).z_T=\sqrt{\bar\alpha_T}\,z_y+\sqrt{1-\bar\alpha_T}\,\epsilon, \qquad \epsilon\sim\mathcal N(0,I).

Here zTz_T has the latent shape expected by the diffusion model. The LR signal replaces an all-noise start with a condition-compatible starting point. This is especially relevant when the pretrained training schedule retains a small terminal signal. It is an inference strategy, not a new likelihood or a guarantee that every hallucination disappears.

Results

The main paper tests ×4 SR on synthetic and captured-pair datasets and a collection of real low-quality images. Its user study reports a 38.6% confusion rate against ground truth for synthetic examples and a 57.1% best-selection rate for real examples. These answer different questions and have different candidate sets; they are not interchangeable accuracy scores.

The supplementary timing experiment uses one 32 GB V100 for 128 × 128 input to 512 × 512 output. SeeSR uses 50 steps and 7.24 seconds, compared with 18.70 seconds for 200-step StableSR and 0.09 seconds for Real-ESRGAN. These are reported system measurements, not predictions for a phone or current GPU. Counting denoiser steps alone also omits the semantic extractor and image encoder.

Ablations

Supplementary Table 1 removes LR embedding. On DRealSR, adding it changes PSNR from 26.55 to 27.90 and LPIPS from 0.3952 to 0.3299, while CLIPIQA decreases from 0.7248 to 0.6708. Better agreement with a reference can therefore lower a predicted-quality score. On synthetic DIV2K, FID slightly worsens from 32.53 to 32.79 even as PSNR improves from 20.58 to 21.04: not every metric improves together.

Table 3 compares the prompt paths. Without DAPE and representation cross-attention, DIV2K FID is 37.35; substituting the original clean-image tagger gives 46.34. Hard prompts alone give 38.92, soft prompts alone 38.77, and the complete model 32.79. Soft-only conditioning has slightly lower LPIPS than the complete model in that table, while visual examples show semantic mistakes. Thus the complementarity claim concerns semantic preservation and the collection of measures, not a win in every table cell. The degraded-image tagging experiment separately supports improved tag prediction, rather than assuming the SR outputs themselves prove tag accuracy.

Conclusion

Reliable semantic conditions need training for the damage they will encounter. Combining structure, tags and continuous features gives the prior more useful guidance than supplying generic clean-image labels.

Problems

DAPE can still misclassify severe damage, and the generative prior can amplify that mistake. Image-to-text and image-feature alignment remain imperfect. Small text and faces are difficult, while the additional tagger costs memory and runtime. LR embedding reduces some unnecessary texture but does not make synthesized detail factual. One-step descendants reuse prompt extraction, so their speed reports must include its cost when it is part of deployment.

References

  1. N. Efrat, D. Glasner, A. Apartsin, B. Nadler and A. Levin, “Accurate Blur Models vs. Image Priors in Single Image Super-Resolution,” in Proc. IEEE ICCV, 2013, pp. 2832–2839. doi · CVF
  2. K. Zhang, W. Zuo and L. Zhang, “Learning a Single Convolutional Super-Resolution Network for Multiple Degradations,” CVPR, 2018. paper
  3. A. Liu, Y. Liu, J. Gu, Y. Qiao, C. Dong, “Blind Image Super-Resolution: A Survey and Beyond,” IEEE TPAMI (early access), 2022. source
  4. Honggang Chen, Xiaohai He, Linbo Qing, Yuanyuan Wu, Chao Ren, Ray E. Sheriff, Ce Zhu, “Real-world single image super-resolution: A brief review,” Information Fusion, 2022. source
  5. B. B. Moser, A. S. Shanbhag, F. Raue, S. Frolov, S. Palacio and A. Dengel, “Diffusion Models, Image Super-Resolution, and Everything: A Survey,” IEEE Transactions on Neural Networks and Learning Systems, vol. 36, no. 7, pp. 11793–11813, 2025. doi · arXiv
  6. Y. Xu, S. R. Tseng, Y. Tseng, H. Kuo, Y. Tsai, “Unified Dynamic Convolutional Network for Super-Resolution with Variational Degradations,” CVPR, 2020. source
  7. K. Zhang, L. Van Gool, R. Timofte, “Deep Unfolding Network for Image Super-Resolution,” CVPR, 2020. source
  8. J. Gu, H. Lu, W. Zuo and C. Dong, “Blind Super-Resolution With Iterative Kernel Correction,” CVPR, 2019. paper
  9. Z. Luo, Y. Huang, S. Li, L. Wang and T. Tan, “Unfolding the Alternating Optimization for Blind Super Resolution,” NeurIPS, 2020. paper
  10. Z. Luo, H. Huang, L. Yu, Y. Li, H. Fan and S. Liu, “Deep Constrained Least Squares for Blind Image Super-Resolution,” CVPR, 2022. paper
  11. T. Michaeli, M. Irani, “Nonparametric Blind Super-resolution,” Proceedings of the IEEE International Conference on Computer Vision, 2013. source
  12. S. Bell-Kligler, A. Shocher and M. Irani, “Blind Super-Resolution Kernel Estimation using an Internal-GAN,” NeurIPS, 2019. paper
  13. A. Shocher, N. Cohen and M. Irani, “‘Zero-Shot’ Super-Resolution Using Deep Internal Learning,” in Proc. IEEE/CVF CVPR, 2018, pp. 3118–3126. doi · arXiv
  14. L. Wang, Y. Wang, X. Dong, Q. Xu, J. Yang, W. An and Y. Guo, “Unsupervised Degradation Representation Learning for Blind Super-Resolution,” CVPR, 2021. paper
  15. B. Xia, Y. Zhang, Y. Wang, Y. Tian, W. Yang, R. Timofte and L. Van Gool, “Knowledge Distillation Based Degradation Estimation for Blind Super-Resolution,” ICLR, 2023. paper
  16. K. Zhang, J. Liang, L. Van Gool, R. Timofte, “Designing a Practical Degradation Model for Deep Blind Image Super-Resolution,” Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021. source
  17. X. Wang, L. Xie, C. Dong, Y. Shan, “Real-ESRGAN: Training Real-World Blind Super-Resolution With Pure Synthetic Data,” Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021. source
  18. J. Liang, H. Zeng, L. Zhang, “Efficient and Degradation-Adaptive Network for Real-World Image Super-Resolution,” ECCV, 2022. source
  19. J. Cai, H. Zeng, H. Yong, Z. Cao, L. Zhang, “Toward Real-World Single Image Super-Resolution: A New Benchmark and a New Model,” Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019. source
  20. P. Wei, Z. Xie, H. Lu, Z. Zhan, Q. Ye, W. Zuo, L. Lin, “Component Divide-and-Conquer for Real-World Image Super-Resolution,” ECCV, 2020. source
  21. C. Chen, Z. Xiong, X. Tian, Z. Zha, F. Wu, “Camera Lens Super-Resolution,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019. source
  22. Y. Yuan, S. Liu, J. Zhang, Y. Zhang, C. Dong, L. Lin, “Unsupervised Image Super-Resolution using Cycle-in-Cycle Generative Adversarial Networks,” CVPR Workshops, 2018. source
  23. A. Bulat, J. Yang, G. Tzimiropoulos, “To learn image super-resolution, use a GAN to learn how to do image degradation first,” ECCV, 2018. source
  24. M. Fritsche, S. Gu, R. Timofte, “Frequency Separation for Real-World Super-Resolution,” ICCV Workshops, 2019. source
  25. X. Ji, Y. Cao, Y. Tai, C. Wang, J. Li, F. Huang, “Real-World Super-Resolution via Kernel Estimation and Noise Injection,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2020. source
  26. J. Wang, Z. Yue, S. Zhou, K. C. K. Chan, C. C. Loy, “Exploiting Diffusion Prior for Real-World Image Super-Resolution,” IJCV, 2024. source
  27. X. Lin, J. He, Z. Chen, et al., “DiffBIR: Towards Blind Image Restoration with Generative Diffusion Prior,” ECCV, 2024. source
  28. T. Yang, R. Wu, P. Ren, X. Xie, L. Zhang, “Pixel-Aware Stable Diffusion for Realistic Image Super-resolution and Personalized Stylization,” ECCV, 2024. source
  29. R. Wu, T. Yang, L. Sun, Z. Zhang, S. Li, L. Zhang, “SeeSR: Towards Semantics-Aware Real-World Image Super-Resolution,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024. source · supplementary experiments
  30. F. Yu, J. Gu, Z. Li, et al., “Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild,” CVPR, 2024. source
  31. Z. Yue, J. Wang, C. C. Loy, “ResShift: Efficient Diffusion Model for Image Super-resolution by Residual Shifting,” NeurIPS, 2023. source
  32. X. Wang, K. Yu, C. Dong, C. C. Loy, “Recovering Realistic Texture in Image Super-Resolution by Deep Spatial Feature Transform,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018. source
  33. K. He, H. Fan, Y. Wu, S. Xie, R. Girshick, “Momentum Contrast for Unsupervised Visual Representation Learning,” CVPR, 2020. source
  34. X. Wang, K. Yu, S. Wu, J. Gu, Y. Liu, C. Dong, Y. Qiao and C. C. Loy, “ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks,” in Proc. ECCV Workshops, 2018 (Lecture Notes in Computer Science, 2019), pp. 63–79. doi · arXiv
  35. O. Ronneberger, P. Fischer and T. Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation,” in Proc. MICCAI, 2015, pp. 234–241. doi · arXiv
  36. T. Miyato, T. Kataoka, M. Koyama, Y. Yoshida, “Spectral Normalization for Generative Adversarial Networks,” ICLR, 2018. source
  37. J. Liang, H. Zeng, L. Zhang, “Details or Artifacts: A Locally Discriminative Learning Approach to Realistic Image Super-Resolution,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022. source
  38. A. Ignatov, N. Kobyshev, R. Timofte, K. Vanhoey, L. Van Gool, “DSLR-Quality Photos on Mobile Devices with Deep Convolutional Networks,” ICCV, 2017. source
  39. A. Lugmayr, M. Danelljan, R. Timofte, “Unsupervised Learning for Real-World Super-Resolution,” arXiv preprint, 2019. source
  40. A. Lugmayr, M. Danelljan, R. Timofte, et al., “NTIRE 2020 Challenge on Real-World Image Super-Resolution: Methods and Results,” CVPR Workshops, 2020. source
  41. R. Rombach, A. Blattmann, D. Lorenz, P. Esser, B. Ommer, “High-Resolution Image Synthesis with Latent Diffusion Models,” CVPR, 2022. source
  42. L. Zhang, A. Rao, M. Agrawala, “Adding Conditional Control to Text-to-Image Diffusion Models,” ICCV, 2023. source
  43. Y. Ai, X. Zhou, H. Huang, X. Han, Z. Chen, Q. You, H. Yang, “DreamClear: High-Capacity Real-World Image Restoration with Privacy-Safe Dataset Curation,” NeurIPS, 2024. source
  44. Z. Duan, J. Zhang, X. Jin, et al., “DiT4SR: Taming Diffusion Transformer for Real-World Image Super-Resolution,” ICCV, 2025. source
  45. R. Wu, L. Sun, Z. Ma, L. Zhang, “One-Step Effective Diffusion Network for Real-World Image Super-Resolution,” NeurIPS, 2024. source
  46. Y. Wang, W. Yang, X. Chen, et al., “SinSR: Diffusion-Based Image Super-Resolution in a Single Step,” CVPR, 2024. source
  47. E. Agustsson and R. Timofte, “NTIRE 2017 Challenge on Single Image Super-Resolution: Dataset and Study,” in Proc. IEEE CVPR Workshops, 2017, pp. 1122–1131. doi · CVF
  48. Z. Wang, A. C. Bovik, H. R. Sheikh and E. P. Simoncelli, “Image Quality Assessment: From Error Visibility to Structural Similarity,” IEEE Transactions on Image Processing, vol. 13, no. 4, pp. 600–612, 2004. doi
  49. R. Zhang, P. Isola, A. A. Efros, E. Shechtman and O. Wang, “The Unreasonable Effectiveness of Deep Features as a Perceptual Metric,” in Proc. IEEE/CVF CVPR, 2018, pp. 586–595. doi · arXiv
  50. K. Ding, K. Ma, S. Wang, E. P. Simoncelli, “Image Quality Assessment: Unifying Structure and Texture Similarity,” IEEE TPAMI, 2022. source
  51. A. Mittal, R. Soundararajan and A. C. Bovik, “Making a “Completely Blind” Image Quality Analyzer,” IEEE Signal Processing Letters, vol. 20, no. 3, pp. 209–212, 2013. doi
  52. J. Ke, Q. Wang, Y. Wang, P. Milanfar, F. Yang, “MUSIQ: Multi-scale Image Quality Transformer,” ICCV, 2021. source
  53. S. Yang, T. Wu, S. Shi, S. Lao, Y. Gong, M. Cao, J. Wang, Y. Yang, “MANIQA: Multi-dimension Attention Network for No-Reference Image Quality Assessment,” CVPR Workshops, 2022. source
  54. J. Wang, K. C. K. Chan, C. C. Loy, “Exploring CLIP for Assessing the Look and Feel of Images,” AAAI, 2023. source
  55. M. Heusel, H. Ramsauer, T. Unterthiner, B. Nessler, S. Hochreiter, “GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium,” NeurIPS, 2017. source
  56. R. Timofte, S. Gu, J. Wu, L. Van Gool, “NTIRE 2018 Challenge on Single Image Super-Resolution: Methods and Results,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2018. source
  57. J. Cai, S. Gu, R. Timofte, L. Zhang, “NTIRE 2019 Challenge on Real Image Super-Resolution: Methods and Results,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2019. source
  58. A. Lugmayr, M. Danelljan, R. Timofte, et al., “AIM 2019 Challenge on Real-World Image Super-Resolution: Methods and Results,” ICCV Workshops, 2019. source
  59. P. Wei, H. Lu, R. Timofte, et al., “AIM 2020 Challenge on Real Image Super-Resolution: Methods and Results,” ECCV Workshops, 2020. source
  60. Z. Chen, K. Liu, J. Gong, J. Wang, L. Sun, Z. Wu, R. Timofte, Y. Zhang et al., “NTIRE 2025 Challenge on Image Super-Resolution (×4): Methods and Results,” in Proc. IEEE/CVF CVPR Workshops, 2025, pp. 1525–1535. doi · arXiv
  61. Le Zhang, Ao Li, Qibin Hou, Ce Zhu, Yonina C. Eldar, “Deep-Learning-Empowered Super Resolution: A Comprehensive Survey and Future Prospects,” Proceedings of the IEEE, 2025. source
  62. W. Ren, R. Goyal, Z. Hu, T. T. Aumentado-Armstrong, I. Mohomed and A. Levinshtein, “Hallucination Score: Towards Mitigating Hallucinations in Generative Image Super-Resolution,” arXiv:2507.14367, 2025. arXiv
  63. Real-ESRGAN contributors, “Real-ESRGAN model and ×4-plus training configuration,” official source, accessed 2026-11-02. model · configuration
  64. D. P. Kingma, J. Ba, “Adam: A Method for Stochastic Optimization,” ICLR, 2015. source
  65. M. Bevilacqua, A. Roumy, C. Guillemot and M.-L. Alberi Morel, “Low-Complexity Single-Image Super-Resolution based on Nonnegative Neighbor Embedding,” in Proc. BMVC, 2012, pp. 135.1–135.10. doi
  66. R. Zeyde, M. Elad and M. Protter, “On Single Image Scale-Up Using Sparse-Representations,” in Curves and Surfaces, Lecture Notes in Computer Science, Springer, 2012, pp. 711–730. doi
  67. D. Martin, C. Fowlkes, D. Tal and J. Malik, “A Database of Human Segmented Natural Images and Its Application to Evaluating Segmentation Algorithms and Measuring Ecological Statistics,” in Proc. IEEE ICCV, 2001, vol. 2, pp. 416–423. doi
  68. J.-B. Huang, A. Singh and N. Ahuja, “Single Image Super-Resolution from Transformed Self-Exemplars,” in Proc. IEEE CVPR, 2015, pp. 5197–5206. doi · CVF
  69. J. Kim, J. K. Lee and K. M. Lee, “Accurate Image Super-Resolution Using Very Deep Convolutional Networks,” in Proc. IEEE CVPR, 2016, pp. 1646–1654. doi · arXiv
  70. C. Ledig, L. Theis, F. Huszár, J. Caballero, A. Cunningham, A. Acosta, A. Aitken, A. Tejani, J. Totz, Z. Wang and W. Shi, “Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network,” in Proc. IEEE CVPR, 2017, pp. 105–114. doi · arXiv
  71. R. Timofte, E. Agustsson, L. Van Gool, M.-H. Yang, L. Zhang et al., “NTIRE 2017 Challenge on Single Image Super-Resolution: Methods and Results,” in Proc. IEEE CVPR Workshops, 2017, pp. 1110–1121. doi · CVF
  72. Y. Matsui, K. Ito, Y. Aramaki, A. Fujimoto, T. Ogawa, T. Yamasaki and K. Aizawa, “Sketch-Based Manga Retrieval Using Manga109 Dataset,” Multimedia Tools and Applications, vol. 76, no. 20, pp. 21811–21838, 2017. doi
  73. KernelGAN contributors, “Adversarial losses and kernel constraints,” official source, accessed 2026-11-02. loss definitions · alternating updates
  74. X. Mao, Q. Li, H. Xie, R. Y. K. Lau, Z. Wang, S. P. Smolley, “Least Squares Generative Adversarial Networks,” ICCV, 2017. source
  75. DASR contributors, “MoCo logits and contrastive cross-entropy implementation,” official DASR source, commit 51524af8021c94a502ee002e135867a1b75dc7d0, accessed 2026-11-02. encoder · loss
  76. Y. Zhang, K. Li, K. Li, L. Wang, B. Zhong and Y. Fu, “Image Super-Resolution Using Very Deep Residual Channel Attention Networks,” in Proc. ECCV, 2018, pp. 294–310. doi · arXiv
  77. K. Simonyan and A. Zisserman, “Very Deep Convolutional Networks for Large-Scale Image Recognition,” in Proc. ICLR, 2015. arXiv
  78. Y. Zhang, X. Huang, J. Ma, et al., “Recognize Anything: A Strong Image Tagging Model,” CVPR Workshops, 2024. source
  79. E. J. Hu, Y. Shen, P. Wallis, Z. Allen-Zhu, Y. Li, S. Wang, L. Wang, W. Chen, “LoRA: Low-Rank Adaptation of Large Language Models,” ICLR, 2022. source
  80. D. Podell, Z. English, K. Lacey, A. Blattmann, T. Dockhorn, J. Müller, J. Penna, R. Rombach, “SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis,” ICLR, 2024. source
  81. H. Liu, C. Li, Q. Wu, Y. J. Lee, “Visual Instruction Tuning,” NeurIPS, 2023. source
  82. T. Karras, M. Aittala, T. Aila, S. Laine, “Elucidating the Design Space of Diffusion-Based Generative Models,” NeurIPS, 2022. source