Pith. sign in

REVIEW 3 major objections 7 minor 1 cited by

FaithDiff: Unleashing Diffusion Priors for Faithful Image Super-resolution

T0 review · 3 major / 7 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Diffusion super-resolution is more faithful when the diffusion prior is fine-tuned jointly with the encoder, not frozen.

desk verdict A plausible diffusion-SR recipe with honest component ablations, but the real-world faithfulness claim outruns the evidence. read the letter →

arxiv 2411.18824 v1 pith:BZHW54XG submitted 2024-11-27 cs.CV

Junyang Chen , Jinshan Pan , Jiangxin Dong This is my paper · ORCID
classification cs.CV
keywords latentdiffusionmodelimagesuper-resolutionfaithfulrestorationjointfine-tuningalignmentmodulepriorstructuralfidelityreal-worldSR
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

FaithDiff claims that diffusion-based image super-resolution is held back when the diffusion model is frozen, because a pre-trained denoiser treats errors in the low-quality image features as real structures and amplifies them. The paper's remedy is to fine-tune the diffusion model together with the encoder that maps the degraded input into latent space, connected by a lightweight alignment module that merges encoder features with the noisy latent at every denoising step. If the claim is correct, the prevailing recipe of keeping the diffusion prior fixed and adding stronger adaptors is backwards: letting the prior adapt to degraded inputs is what allows faithful recovery of fine structures while keeping results realistic. The paper backs this with better perceptual scores than state-of-the-art diffusion SR methods and with an OCR test showing that restored text stays readable.

What carries the argument

The load-bearing mechanism is the joint fine-tuning loop. The alignment module computes $f^a_t = \mathrm{Linear}(\mathrm{Trans}(\mathrm{Concat}(\mathrm{Conv}(x^{HQ}_t), \mathrm{Conv}(f^{LQ}))) + \mathrm{Conv}(x^{HQ}_t))$ and injects it into each denoising step, while the whole trainable stack—LQ encoder, alignment module, and denoiser—is optimized on $\lVert \epsilon - \hat{\epsilon}_\theta(\sqrt{\bar\alpha_t}x^{HQ}_0 + \sqrt{1-\bar\alpha_t}\epsilon, f^{LQ}, c, t)\rVert_1$. Because the noisy latent becomes clearer as denoising progresses while the LQ features stay fixed, the alignment module keeps the two streams compatible at every step; the unified optimization is what prevents the denoiser from amplifying encoder mistakes into false structures.

What would settle it

Evaluate FaithDiff on real-world degraded images whose degradations lie far outside the synthetic training distribution—for example, spatially varying motion blur combined with heavy JPEG compression and sensor noise—and compare perceptual scores and OCR accuracy against SeeSR and SUPIR. If the MUSIQ/CLIPIQA+ lead shrinks to noise or reverses on such out-of-distribution inputs, the claim that joint fine-tuning yields faithful real-world structures would be unsupported.

Watch

Extended reading notes

Core claim

In latent-diffusion super-resolution, the encoder that turns a low-quality image into latent features and the denoising network should be optimized as one system, not deployed as a frozen prior. FaithDiff takes the SDXL VAE encoder's penultimate-layer features (512 channels, richer than the 8-channel last layer used by DiffBIR and SUPIR), pushes them together with the noisy latent through an alignment module built from two Transformer blocks, and optimizes the encoder, alignment module, and denoiser jointly on an L1 noise-prediction loss. The paper demonstrates through ablations that freezing either module, or fine-tuning them separately, lowers perceptual quality (MUSIQ 69.7–71.1 versus 72.7 on RealPhoto60) and produces structures inconsistent with the input. Joint optimization is what lets the network separate degradation effects from genuine image content, producing state-of-the-art results on synthetic benchmarks and on real photos, film stills, and social-media images.

Load-bearing premise

The paper assumes that the synthetic degradation model used to create training low-quality images—the same configuration as PASD—is representative enough of real-world degradations that FaithDiff's gains on old photos, film stills, and social-media images are genuine, even though only 238 real images are used for validation.

Editorial extensions

If this is right

  • Fine-tuning the diffusion model instead of freezing it suppresses the failure mode that makes DiffBIR, SeeSR, and SUPIR produce wrong characters and spurious structures.
  • Jointly optimizing the encoder and the diffusion model beats optimizing either one alone: separate fine-tuning scores 69.94 MUSIQ on RealPhoto60 versus 72.74 for joint optimization.
  • FaithDiff restores text more faithfully: OCR precision and recall on restored road signs reach 36.45% and 46.74%, compared with 24.32% and 25.14% for SeeSR and 22.45% and 23.50% for SUPIR.
  • The alignment module removes the need for a ControlNet-style adaptor, cutting diffusion runtime to 2.55 seconds per 1024×1024 image, the fastest among the compared diffusion SR methods.
  • Taking features from the penultimate encoder layer rather than the compressed last layer improves LPIPS on DIV2K-Val by 0.0222.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If joint fine-tuning is what unlocks faithfulness, the same principle likely transfers to other latent-diffusion inverse problems such as deblurring, inpainting, and face restoration, where frozen-prior adaptor designs still dominate; re-running FaithDiff's FT EN & Fix DM ablation in those tasks would test this directly.
  • The OCR-as-faithfulness evaluation offers a cheap, objective proxy for structural fidelity; adopting it as a standard benchmark would help the SR field distinguish 'realistic-looking' from 'content-preserving'.
  • Fine-tuning the VAE encoder could drift its latent space away from what the frozen VAE decoder expects; measuring FaithDiff's encoder-decoder reconstruction error on clean high-quality inputs would reveal whether that drift is paid for elsewhere.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 7 minor

Summary. The manuscript presents FaithDiff, a latent diffusion model (LDM) based super-resolution method that, unlike prior work freezing the diffusion prior, jointly fine-tunes the VAE encoder, a proposed alignment module, and the diffusion model. The alignment module injects penultimate-layer LQ encoder features into the noisy latent via transformer blocks and a linear projection. Experiments are reported on synthetic DIV2K and LSDIR benchmarks, the real-world RealPhoto60 and a self-collected RealDeg set, plus an OCR recognition test. The claimed contribution is that unleashing the diffusion prior through unified optimization yields faithful, structure-preserving SR results that outperform state-of-the-art methods.

Significance. If the central claim is substantiated, the paper makes a useful contribution by showing that fine-tuning, rather than freezing, the diffusion prior can improve fidelity in LDM-based SR, and the simple alignment module plus joint optimization recipe is plausible and well ablated. The ablations in Tables 5 and 6 are honest comparisons with the same training settings, and the run-time advantage in Table 4 is a concrete strength. However, the evidence for the real-world faithfulness claim is incomplete: no-reference metrics and a synthetic-degradation OCR test do not establish structure preservation on genuine real-world degradations. The contribution is therefore potentially significant but requires additional validation.

major comments (3)
  1. [4.2, Table 1] The claim in the abstract and Section 3.3 that FaithDiff 'outperforms state-of-the-art methods' is not uniformly supported by Table 1. On DIV2K-Val Level-I, SUPIR achieves lower LPIPS (0.2139 vs. 0.2187) and DreamClear achieves higher CLIPIQA+ (0.5478 vs. 0.5432); on LSDIR-Val Level-I, SUPIR achieves higher PSNR (21.58 vs. 21.20) and SSIM (0.5961 vs. 0.5760). Please restrict the claim to the perceptual metrics and severe-degradation settings where the advantage holds, or add statistical significance testing to support a broad claim.
  2. [4.2, Table 2 and 4.3] The 'faithful structures' conclusion for real-world images is not established. The real-world evaluation relies solely on no-reference metrics (MUSIQ, CLIPIQA+) over RealPhoto60 and the 238-image RealDeg set, which can reward plausible textures that are not faithful. The OCR test in Section 4.3 generates LQ images with the same synthetic degradation model used in training (Section 4.1), so it does not test generalization to real-world degradations. Please add a reference-based evaluation on a real-world SR benchmark with ground truth, or a human fidelity study, to support the faithfulness claim.
  3. [5, Tables 5-6] The ablations are informative but report only point estimates of MUSIQ and LPIPS without variance or significance tests. For instance, in Table 6 the jointly optimized model's MUSIQ gain over 'FT EN & DM (SP)' is 2.80 on RealPhoto60, while the gap to 'Fix EN & FT DM' is only 1.63. Since training is stochastic, please report standard deviations over repeated runs or otherwise justify that the observed differences are not noise.
minor comments (7)
  1. [4.3] The text 'following the same method as described in Seciton 4.1' contains a typo: 'Seciton' should be 'Section'.
  2. [Table 4] The header 'DreamClear [41]' is incorrect; DreamClear is reference [2], while [41] is SeeSR.
  3. [Table 2] The entry 'StabeSR' should be spelled 'StableSR'.
  4. [Figure 2 caption] The phrase 'high viusal quality' should be corrected to 'high visual quality'.
  5. [5.1] The word 'latant' in 'the noisy latant of the diffusion model' should be 'latent'.
  6. [4.1] The degradation configuration is only cited as '[43]'; please specify the exact degradation model and parameter ranges so that the training setup is reproducible.
  7. [3.3, Eq. (3)] Equation (3) has inconsistent spacing and the norm delimiters are not clearly closed in the formatted text; please fix the LaTeX rendering.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: FaithDiff is trained with a standard diffusion noise-prediction loss on external data and validated by ablations and benchmark comparisons, not by fitting the claimed outcome.

full rationale

The derivation chain is self-contained. The training objective in Eq. (3) is a conventional L1 noise-prediction loss over synthetic LQ-HQ pairs, and the design choices (alignment module, penultimate-layer features, unified fine-tuning) are justified by ablations in Tables 5 and 6 against held-out metrics rather than by optimizing those metrics. No parameter is fitted to the reported evaluation target, and no claim reduces by construction to its own input. The paper contains no load-bearing self-citation; references such as [43] and [44] are external baselines and degradation recipes. The only potentially self-referential element is the OCR test in Section 4.3, where LQ images are generated 'using the same method as described in Section 4.1'—the same degradation pipeline used for training. That weakness makes the OCR result weaker evidence for real-world faithfulness, but it is an evaluation limitation, not a circular derivation: the restored images are produced by the trained model and then measured by an external OCR recognizer, so the model's parameters do not reduce to the OCR metric. Real-world evaluation on RealPhoto60 and RealDeg uses external no-reference metrics and comparisons with prior methods, so the central claim remains independent of its inputs.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No new physical entities or fitted constants are introduced. The alignment module is a new network component, not a postulated entity. The listed axioms are the load-bearing assumptions about synthetic-to-real transfer, metric validity, and the frozen decoder's stability.

assumptions (4)
  • domain assumption The synthetic degradation model following PASD adequately represents real-world degradations.
    Used to generate LQ training data in Section 4.1; the real-world transfer of the method depends on this assumption.
  • domain assumption MUSIQ and CLIPIQA+ are valid proxies for faithful image quality in the absence of ground truth.
    Used as the main metrics on RealPhoto60 and RealDeg in Section 4.2, Table 2.
  • domain assumption OCR accuracy on synthetically degraded road text reflects structural fidelity.
    Used in Section 4.3 to support the 'faithful structures' claim; the degradations are synthesized with the same model as training, not real occlusions.
  • domain assumption The SDXL VAE decoder remains a strong prior when the encoder and UNet are fine-tuned.
    The decoder is frozen in Section 3.3, and the method assumes it can reconstruct HQ images from the modified latents.

how reviews work

0 comments
Cite this review

Pith. "Pith review of FaithDiff: Unleashing Diffusion Priors for Faithful Image Super-resolution." pith.science (2026). https://pith.science/paper/BZHW54XG

@misc{pith2026241118824,
  author       = {Pith},
  title        = {Pith review of: FaithDiff: Unleashing Diffusion Priors for Faithful Image Super-resolution},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BZHW54XG}},
  note         = {Machine review of arXiv:2411.18824}
}
read the original abstract

Faithful image super-resolution (SR) not only needs to recover images that appear realistic, similar to image generation tasks, but also requires that the restored images maintain fidelity and structural consistency with the input. To this end, we propose a simple and effective method, named FaithDiff, to fully harness the impressive power of latent diffusion models (LDMs) for faithful image SR. In contrast to existing diffusion-based SR methods that freeze the diffusion model pre-trained on high-quality images, we propose to unleash the diffusion prior to identify useful information and recover faithful structures. As there exists a significant gap between the features of degraded inputs and the noisy latent from the diffusion model, we then develop an effective alignment module to explore useful features from degraded inputs to align well with the diffusion process. Considering the indispensable roles and interplay of the encoder and diffusion model in LDMs, we jointly fine-tune them in a unified optimization framework, facilitating the encoder to extract useful features that coincide with diffusion process. Extensive experimental results demonstrate that FaithDiff outperforms state-of-the-art methods, providing high-quality and faithful SR results.

Figures

Figures reproduced from arXiv: 2411.18824 by the authors.

Figure 1
Figure 1. Visual comparison with state-of-the-art SR methods. (b) and (c) are the intermediate results of the degradation removal module [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. An overview of the FaithDiff, which takes LQ images and image descriptions as inputs and restores HQ images via diffusion process. To fully leverage the power of LDMs, we propose to unleash diffusion priors. An alignment module is developed to effectively incorporate the features extracted from LQ images with the noisy latent of the diffusion model. We jointly optimize the encoder, the alignment module, and the diff… view at source ↗
Figure 3
Figure 3. Image SR result (×4) on the synthetic benchmark. The restored image by GAN-based methods [ [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Image SR result (×2) on the real-world benchmarks. Compared to competing methods, our approach generates more realistic [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 6
Figure 6. Figure 6: Visualization of DAAMs [33] for ‘bottles’. (b)-(d) are DAAMs for PASD [43], SeeSR [41], and our method [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 5
Figure 5. Figure 5: Effectiveness of the unified feature optimization on im [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Text-Aware Image Restoration with Diffusion Models

    cs.CV 2025-06 conditional novelty 6.0 of 10

    A diffusion restoration model jointly trained with a text-spotting module and prompted by its own recognized text improves text recognition accuracy on restored images compared with general-purpose restoration methods.

Reference graph

Works this paper leans on

49 extracted references · 30 canonical work pages · cited by 1 Pith paper

  1. [1]

    Ntire 2017 challenge on single image super-resolution: Dataset and study

    Eirikur Agustsson and Radu Timofte. Ntire 2017 challenge on single image super-resolution: Dataset and study. In CVPR Workshops, 2017. 4, 5, 7, 8

    work page 2017

  2. [2]

    Dream- clear: High-capacity real-world image restoration with privacy-safe dataset curation

    Yuang Ai, Xiaoqiang Zhou, Huaibo Huang, Xiaotian Han, Zhengyu Chen, Quanzeng You, and Hongxia Yang. Dream- clear: High-capacity real-world image restoration with privacy-safe dataset curation. In NeurIPS, 2024. 5, 6, 7

    work page 2024

  3. [3]

    Glean: Generative latent bank for large- factor image super-resolution

    Kelvin CK Chan, Xintao Wang, Xiangyu Xu, Jinwei Gu, and Chen Change Loy. Glean: Generative latent bank for large- factor image super-resolution. In CVPR, 2021. 1

    work page 2021

  4. [4]

    Real-world blind super-resolution via feature matching with implicit high- resolution priors

    Chaofeng Chen, Xinyu Shi, Yipeng Qin, Xiaoming Li, Xi- aoguang Han, Tao Yang, and Shihui Guo. Real-world blind super-resolution via feature matching with implicit high- resolution priors. In ACM MM, 2022. 1, 2

    work page 2022

  5. [5]

    Pixart-α: Fast training of diffusion trans- former for photorealistic text-to-image synthesis

    Junsong Chen, Jincheng Yu, Chongjian Ge, Lewei Yao, Enze Xie, Yue Wu, Zhongdao Wang, James Kwok, Ping Luo, Huchuan Lu, et al. Pixart-α: Fast training of diffusion trans- former for photorealistic text-to-image synthesis. In ICLR,

  6. [6]

    Blind super-resolution with iterative kernel correction

    Jinjin Gu, Hannan Lu, Wangmeng Zuo, and Chao Dong. Blind super-resolution with iterative kernel correction. In CVPR, 2019. 2

    work page 2019

  7. [7]

    Div8k: Diverse 8k resolution image dataset

    Shuhang Gu, Andreas Lugmayr, Martin Danelljan, Manuel Fritsche, Julien Lamour, and Radu Timofte. Div8k: Diverse 8k resolution image dataset. In ICCV Workshops, 2019. 4

    work page 2019

  8. [8]

    Classifier-free diffusion guidance

    Jonathan Ho and Tim Salimans. Classifier-free diffusion guidance. arXiv preprint arXiv:2207.12598, 2022. 5

    arXiv 2022

Show all 49 references
  1. [9]

    Denoising diffu- sion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffu- sion probabilistic models. In NeurIPS, 2020. 2

    2020

  2. [10]

    Animate anyone: Consistent and controllable image-to-video synthesis for character animation

    Li Hu, Xin Gao, Peng Zhang, Ke Sun, Bang Zhang, and Liefeng Bo. Animate anyone: Consistent and controllable image-to-video synthesis for character animation. In CVPR,

  3. [11]

    Un- folding the alternating optimization for blind super resolu- tion

    Yan Huang, Shang Li, Liang Wang, Tieniu Tan, et al. Un- folding the alternating optimization for blind super resolu- tion. In NeurIPS, 2020. 2

    2020

  4. [12]

    A style-based generator architecture for generative adversarial networks

    Tero Karras, Samuli Laine, and Timo Aila. A style-based generator architecture for generative adversarial networks. In CVPR, 2019. 4

    2019

  5. [13]

    Elucidating the design space of diffusion-based generative models

    Tero Karras, Miika Aittala, Timo Aila, and Samuli Laine. Elucidating the design space of diffusion-based generative models. In NeurIPS, 2022. 5

    2022

  6. [14]

    Musiq: Multi-scale image quality transformer

    Junjie Ke, Qifei Wang, Yilin Wang, Peyman Milanfar, and Feng Yang. Musiq: Multi-scale image quality transformer. In CVPR, 2021. 5, 6

    2021

  7. [15]

    Auto-encoding variational bayes

    Diederik P Kingma. Auto-encoding variational bayes. In ICLR, 2014. 3

    2014

  8. [16]

    Pp-ocrv3: More attempts for the im- provement of ultra lightweight ocr system

    Chenxia Li, Weiwei Liu, Ruoyu Guo, Xiaoting Yin, Kaitao Jiang, Yongkun Du, Yuning Du, Lingfeng Zhu, Baohua Lai, Xiaoguang Hu, et al. Pp-ocrv3: More attempts for the im- provement of ultra lightweight ocr system. arXiv preprint arXiv:2206.03001, 2022. 6

    2022 arXiv

  9. [17]

    Lsdir: A large scale dataset for image restoration

    Yawei Li, Kai Zhang, Jingyun Liang, Jiezhang Cao, Ce Liu, Rui Gong, Yulun Zhang, Hao Tang, Yun Liu, Denis Deman- dolx, et al. Lsdir: A large scale dataset for image restoration. In CVPR, 2023. 4, 5, 6

    2023

  10. [18]

    Efficient and degradation-adaptive network for real-world image super- resolution

    Jie Liang, Hui Zeng, and Lei Zhang. Efficient and degradation-adaptive network for real-world image super- resolution. In ECCV, 2022. 2, 4

    2022

  11. [19]

    Details or artifacts: A locally discriminative learning approach to realistic image super-resolution

    Jie Liang, Hui Zeng, and Lei Zhang. Details or artifacts: A locally discriminative learning approach to realistic image super-resolution. In CVPR, 2022. 1, 2

    2022

  12. [20]

    Enhanced deep residual networks for single image super-resolution

    Bee Lim, Sanghyun Son, Heewon Kim, Seungjun Nah, and Kyoung Mu Lee. Enhanced deep residual networks for single image super-resolution. In CVPR Workshops, 2017. 4

    2017

  13. [21]

    Diff- bir: Towards blind image restoration with generative diffu- sion prior

    Xinqi Lin, Jingwen He, Ziyan Chen, Zhaoyang Lyu, Bo Dai, Fanghua Yu, Wanli Ouyang, Yu Qiao, and Chao Dong. Diff- bir: Towards blind image restoration with generative diffu- sion prior. arXiv preprint arXiv:2308.15070, 2023. 1, 2, 3, 5, 6, 7

    2023 arXiv

  14. [22]

    Visual instruction tuning

    Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. Visual instruction tuning. In NeurIPS, 2024. 4

    2024

  15. [23]

    Sgdr: Stochastic gradient descent with warm restarts

    Ilya Loshchilov and Frank Hutter. Sgdr: Stochastic gradient descent with warm restarts. In ICLR, 2017. 5

    2017

  16. [24]

    Decoupled weight decay regularization

    Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization. In ICLR, 2019. 4

    2019

  17. [25]

    T2i-adapter: Learning adapters to dig out more controllable ability for text-to-image diffusion models

    Chong Mou, Xintao Wang, Liangbin Xie, Yanze Wu, Jian Zhang, Zhongang Qi, and Ying Shan. T2i-adapter: Learning adapters to dig out more controllable ability for text-to-image diffusion models. In AAAI, 2024. 2

    2024

  18. [26]

    Sdxl: Improving latent diffusion models for high-resolution image synthesis

    Dustin Podell, Zion English, Kyle Lacey, Andreas Blattmann, Tim Dockhorn, Jonas M ¨uller, Joe Penna, and Robin Rombach. Sdxl: Improving latent diffusion models for high-resolution image synthesis. In ICLR, 2024. 1, 2, 4

    2024

  19. [27]

    Xpsr: Cross-modal priors for diffusion-based image super-resolution

    Yunpeng Qu, Kun Yuan, Kai Zhao, Qizhi Xie, Jinhua Hao, Ming Sun, and Chao Zhou. Xpsr: Cross-modal priors for diffusion-based image super-resolution. In ECCV, 2024. 1, 2

    2024

  20. [28]

    Learn- ing transferable visual models from natural language super- vision

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, et al. Learn- ing transferable visual models from natural language super- vision. In ICML, 2021. 4

    2021

  21. [29]

    Hierarchical text-conditional image gen- eration with clip latents

    Aditya Ramesh, Prafulla Dhariwal, Alex Nichol, Casey Chu, and Mark Chen. Hierarchical text-conditional image gen- eration with clip latents. arXiv preprint arXiv:2204.06125,

    arXiv

  22. [30]

    High-resolution image syn- thesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj¨orn Ommer. High-resolution image syn- thesis with latent diffusion models. In CVPR, 2022. 2

    2022

  23. [31]

    High-resolution image syn- thesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj¨orn Ommer. High-resolution image syn- thesis with latent diffusion models. In CVPR, 2022. 1, 2

    2022

  24. [32]

    Photorealistic text-to-image diffusion models with deep language understanding

    Chitwan Saharia, William Chan, Saurabh Saxena, Lala Li, Jay Whang, Emily L Denton, Kamyar Ghasemipour, Raphael Gontijo Lopes, Burcu Karagol Ayan, Tim Salimans, et al. Photorealistic text-to-image diffusion models with deep language understanding. In NeurIPS, 2022. 1, 2

    2022

  25. [33]

    What the daam: Interpreting stable diffu- sion using cross attention

    Raphael Tang, Linqing Liu, Akshat Pandey, Zhiying Jiang, Gefei Yang, Karun Kumar, Pontus Stenetorp, Jimmy Lin, and Ferhan Ture. What the daam: Interpreting stable diffu- sion using cross attention. arXiv preprint arXiv:2210.04885,

    arXiv

  26. [34]

    Emo: Emote portrait alive - generating expressive portrait videos with audio2video diffusion model under weak conditions

    Linrui Tian, Qi Wang, Bang Zhang, and Liefeng Bo. Emo: Emote portrait alive - generating expressive portrait videos with audio2video diffusion model under weak conditions. In ECCV, 2024. 3

    2024

  27. [35]

    George Tom, Minesh Mathew, Ajoy Mondal, Dimosthenis Karatzas, C. V . Jawahar, and Jerod Weinman. Icdar2024 challenge on occluded roadtext. In ICDAR2024 Workshops,

  28. [36]

    Attention is all you need

    A Vaswani. Attention is all you need. In NeurIPS, 2017. 4

    2017

  29. [37]

    Ex- ploring clip for assessing the look and feel of images

    Jianyi Wang, Kelvin CK Chan, and Chen Change Loy. Ex- ploring clip for assessing the look and feel of images. In AAAI, 2023. 5, 6

    2023

  30. [38]

    Exploiting diffusion prior for real-world image super-resolution

    Jianyi Wang, Zongsheng Yue, Shangchen Zhou, Kelvin CK Chan, and Chen Change Loy. Exploiting diffusion prior for real-world image super-resolution. IJCV, 2024. 2, 5, 6

    2024

  31. [39]

    To- wards real-world blind face restoration with generative facial prior

    Xintao Wang, Yu Li, Honglun Zhang, and Ying Shan. To- wards real-world blind face restoration with generative facial prior. In CVPR, 2021. 1

    2021

  32. [40]

    Real-esrgan: Training real-world blind super-resolution with pure synthetic data

    Xintao Wang, Liangbin Xie, Chao Dong, and Ying Shan. Real-esrgan: Training real-world blind super-resolution with pure synthetic data. In CVPR, 2021. 1, 2, 5, 6, 7

    2021

  33. [41]

    Seesr: Towards semantics-aware real-world image super-resolution

    Rongyuan Wu, Tao Yang, Lingchen Sun, Zhengqiang Zhang, Shuai Li, and Lei Zhang. Seesr: Towards semantics-aware real-world image super-resolution. In CVPR, 2024. 2, 5, 6, 7, 8

    2024

  34. [42]

    Desra: detect and delete the artifacts of gan-based real-world super-resolution models

    Liangbin Xie, Xintao Wang, Xiangyu Chen, Gen Li, Ying Shan, Jiantao Zhou, and Chao Dong. Desra: detect and delete the artifacts of gan-based real-world super-resolution models. In ICML, 2023. 1, 2

    2023

  35. [43]

    Pixel-aware stable diffusion for realistic im- age super-resolution and personalized stylization

    Tao Yang, Rongyuan Wu, Peiran Ren, Xuansong Xie, and Lei Zhang. Pixel-aware stable diffusion for realistic im- age super-resolution and personalized stylization. In ECCV,

  36. [44]

    Scaling up to excellence: Practicing model scaling for photo- realistic image restoration in the wild

    Fanghua Yu, Jinjin Gu, Zheyuan Li, Jinfan Hu, Xiangtao Kong, Xintao Wang, Jingwen He, Yu Qiao, and Chao Dong. Scaling up to excellence: Practicing model scaling for photo- realistic image restoration in the wild. In CVPR, 2024. 1, 2, 3, 4, 5, 6, 7, 8

    2024

  37. [45]

    Learning a single convolutional super-resolution network for multiple degradations

    Kai Zhang, Wangmeng Zuo, and Lei Zhang. Learning a single convolutional super-resolution network for multiple degradations. In CVPR, 2018. 2

    2018

  38. [46]

    Designing a practical degradation model for deep blind image super-resolution

    Kai Zhang, Jingyun Liang, Luc Van Gool, and Radu Timo- fte. Designing a practical degradation model for deep blind image super-resolution. In CVPR, 2021. 1, 2, 5, 6

    2021

  39. [47]

    Adding conditional control to text-to-image diffusion models

    Lvmin Zhang, Anyi Rao, and Maneesh Agrawala. Adding conditional control to text-to-image diffusion models. In ICCV, 2023. 2, 6, 8

    2023

  40. [48]

    The unreasonable effectiveness of deep features as a perceptual metric

    Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shechtman, and Oliver Wang. The unreasonable effectiveness of deep features as a perceptual metric. In CVPR, 2018. 5

    2018

  41. [49]

    Towards robust blind face restora- tion with codebook lookup transformer

    Shangchen Zhou, Kelvin Chan, Chongyi Li, and Chen Change Loy. Towards robust blind face restora- tion with codebook lookup transformer. In NeurIPS, 2022. 1 10

    2022

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.