Degradation-Aware Image Enhancement via Vision-Language Classification

Jiaming Ding; Jiang Li; Jie Cai; Jinglin Shen; Kangning Yang; Lan Fu; Ling Ouyang; Zibo Meng

REVIEW 4 major objections 4 minor 1 cited by

Degradation-Aware Image Enhancement via Vision-Language Classification

T0 review · 4 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read The paper claims a zero-shot vision-language model can classify image degradation types and route each photo to a specialist restoration model, producing visibly improved results without retraining.

desk verdict A practical VLM-routing idea with no experiment to back it; the central accuracy claim is unsupported by the three example images in Section IV. read the letter →

arxiv 2506.05450 v1 pith:L5ZF7EOI submitted 2025-06-05 cs.CV

Jie Cai , Kangning Yang , Jiaming Ding , Lan Fu , Ling Ouyang , Jiang Li , Jinglin Shen , Zibo Meng This is my paper

classification cs.CV

keywords imagerestorationvision-languagemodelzero-shotclassificationdegradationsuper-resolutionreflectionremovalmotiondeblurring

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

The paper proposes a fully automated image-enhancement pipeline in which a vision-language model classifies every input photo into one of four categories: super-resolution degradation, reflection artifacts, motion blur, or no visible degradation. Based on that label, the image is sent to a restoration model built for the corresponding problem, such as a diffusion-based super-resolver, an inpainting model for reflections, or a deblurring network. The authors argue that this design replaces manual selection of restoration algorithms and scales to real-world use because the classifier needs no fine-tuning. The stated experimental conclusion is that the approach achieves accurate degradation classification and clear visual improvement across all three degradation types.

What carries the argument

The load-bearing component is the zero-shot classification step, in which Qwen2.5-VL receives a fixed textual prompt and outputs one of four letters, A through D, as a routing decision. That decision determines which specialist model handles the image: InvSR with Real-ESRGAN text fusion for super-resolution, a YOLO/YOSO mask plus LaMa inpainting and NAFNet refinement for reflections, and NAFNet deblurring with CodeFormer face restoration for motion blur. The VLM's classification is what turns a collection of independent restoration models into an automated pipeline.

What would settle it

Take a labeled corpus of degraded images covering the four categories, run the paper's exact prompt through Qwen2.5-VL, and compute a confusion matrix; if classification accuracy is close to chance, or if images routed by the classifier show no improvement over the unprocessed input on a perceptual or reference-based metric, the central claim fails.

Watch

Extended reading notes

Core claim

The central claim, stated on the paper's own terms, is that Qwen2.5-VL can act as a zero-shot degradation classifier that reliably separates images into four categories, and that routing each category to a dedicated restoration model yields higher-quality outputs than a one-size-fits-all approach. For super-resolution degradation the pipeline combines InvSR with PaddleOCR-detected text regions restored by Real-ESRGAN; for reflection artifacts it detects strong reflections with YOLO and YOSO, inpaints them with LaMa, and refines weak reflections with NAFNet; for motion blur it deblurs with NAFNet and enhances faces with CodeFormer. The paper presents visual examples as evidence and concludes that the framework is scalable, automated, and effective for real-world image enhancement.

Load-bearing premise

The entire pipeline rests on the assumption that the Qwen2.5-VL vision-language model, used without any task-specific training, will put real-world photos into the correct one of four degradation categories from the prompt alone.

Editorial extensions

If this is right

If the VLM classifies accurately, users no longer need to diagnose an image's defect or choose a restoration algorithm manually, because the same prompt-driven model makes that choice.
The framework is plug-and-play in the sense that the classifier is used without fine-tuning, so new degradation categories could in principle be added by editing the prompt and attaching another specialist model.
The modular design means each restoration stage can be upgraded independently, so improvements in super-resolution, reflection removal, or deblurring should transfer directly into the pipeline.
The authors state that the three enhancement modules are already deployed in consumer smartphones and process tens of thousands of user images daily, implying the approach is practical at scale.

Reading between the lines

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

Editorial inference: the four-way taxonomy puts noise, blur, and JPEG compression under a single label, so an image with only mild compression will be processed by a heavy super-resolution model even though a lighter correction might suffice.
Editorial inference: because the VLM is asked to return a single letter, it has no built-in way to signal uncertainty or mixed degradations; a confidence score or a multi-label output could let the pipeline skip restoration or call two specialists in sequence.
Editorial inference: a quantitative evaluation of the framework could compare end-to-end outputs against the unprocessed input on perceptual metrics or user studies; the paper's visual examples leave the magnitude of the improvement unspecified.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Colleague,

This is a system paper with a nice hook and almost no experimental content. The hook: use a zero-shot VLM (Qwen2.5-VL) to label an image into one of four degradation buckets, then route it to a specialist restoration model. That specific routing idea is new as far as I know, and it is genuinely practical for a phone camera pipeline. The writing is clear, the choice of off-the-shelf models is sensible, and the authors know the literature. The deployment claim (tens of thousands of images a day on OPPO phones) is at least a sign that the system works in practice.

The soft spot is not a soft spot; it is the load-bearing floor. Section IV is three example images and a paragraph saying 'the results show.' There is no classification accuracy, no confusion matrix, no PSNR/LPIPS, no comparison with running a single generic restoration model, no user study. The abstract and introduction claim 'high accuracy' and 'significant improvements' without reporting either measurement. That is a claim without derivation.

Also worth noting: the four categories are not clean. Category A is a union of noise, blur, and JPEG compression; category C is motion blur. A blurry image could legitimately go to either A or C, and the VLM is asked to make that call with no calibration data. There is no discussion of what happens when the VLM is wrong. For a degraded photo, misrouting can actively make it worse. The absence of any error analysis means the central premise is untested.

I agree with the reader's take. This is not a bad idea and not a bad engineering description; it is an unfinished paper. If the authors came back with a proper evaluation — classification accuracy on a few hundred labeled examples, end-to-end quality metrics, and an ablation on misrouting cost — I would look at it again seriously. As it stands, the experiments section is a placeholder.

Who is this for? Someone surveying VLM applications for image restoration, or someone building a phone enhancement pipeline. A serious referee would not be able to verify any claim from the current text, so I would not send it out in its current form. But I would not burn the paper; I would invite a resubmission with real numbers.

Recommendation: reject, but make clear that the bottleneck is the missing evaluation, not the approach.

Referee Report

4 major / 4 minor

Summary. The paper proposes an automated image enhancement pipeline in which a Vision-Language Model (Qwen2.5-VL) classifies an input image into one of four degradation categories — super-resolution degradation (noise, blur, JPEG), reflection artifacts, motion blur, or no visible degradation — and then routes the image to specialized restoration models: InvSR with Real-ESRGAN for text regions (category A), YOLO/YOSO with LaMa and NAFNet for reflections (category B), and NAFNet with CodeFormer for motion blur (category C). The authors state in the abstract, introduction, and conclusion that experiments demonstrate high classification accuracy and significant visual improvement. Section IV, however, contains only three anecdotal example images and no quantitative metrics of any kind.

Significance. If validated, the framework could be a practically useful industrial system, combining a zero-shot VLM classifier with specialist restorers to avoid manual selection of restoration algorithms. The paper has the merit of assembling existing pretrained components into a concrete, deployable pipeline, and the authors state that the modules have been deployed in OPPO smartphones. The central claim, however, is entirely unquantified: the paper reports no classification accuracy, no confusion matrix, no end-to-end image quality metrics (PSNR, SSIM, LPIPS, FID, or user study), and no comparison against baselines or against simply applying each restoration model unconditionally. Because misclassification actively reroutes an image to the wrong restoration model, the untested VLM classification accuracy is load-bearing, and the paper's experimental section does not support the claimed effectiveness.

major comments (4)

[Section IV] The Experiments section contains no quantitative evaluation whatsoever. It reports only three representative images per degradation type in Fig. 5, with no classification accuracy, no confusion matrix, no PSNR/SSIM/LPIPS/FID, and no user study. The sentences 'Experimental results demonstrate...' and 'The results show that our method is effective...' are therefore unsupported. This is load-bearing because the abstract and conclusion both assert high classification accuracy and significant visual improvement, and the paper's contribution is exactly the routing decision made by the VLM.
[Section III.A] The zero-shot VLM classification is the central premise of the pipeline, yet its reliability is untested. The prompt categories are ambiguous and overlapping: category A includes 'blur' generally while category C is 'motion blur'; category D requires distinguishing 'no visible degradation' from subtle real-world artifacts. The paper gives no accuracy measurement, no failure analysis, and no discussion of the downstream cost of misrouting. A wrong label sends the image to the wrong specialist model, potentially degrading an image that was already acceptable, so the absence of any classification evaluation leaves the mechanism unvalidated.
[Section IV, Fig. 5] Even if classification were perfect, the paper does not show that chaining the specialist models preserves or improves quality. InvSR, Real-ESRGAN, LaMa, NAFNet, and CodeFormer are each pretrained for their own tasks; combining them (e.g., text-region fusion in Section III.B, inpainting followed by NAFNet in Section III.C, and deblurring followed by face restoration in Section III.D) can introduce artifacts or alter color and texture. No end-to-end quality metric or comparison to the unmodified specialist models is reported, so the claimed 'significant improvements in visual quality' are not demonstrated.
[Section V] The deployment claim — 'these three AI-powered enhancement modules have already been deployed in OPPO AI smartphones, where they process tens of thousands of user images daily' — is accompanied only by a YouTube link and no on-device accuracy or quality data. This statement cannot substitute for the missing experimental evaluation, and it is not verifiable from the manuscript. If deployment data exist, they should be reported as measurements; otherwise, the claim should be removed or clearly labeled as an industrial deployment note rather than evidence.

minor comments (4)

[Section III.A] The category definitions should be clarified to avoid overlap, particularly the boundary between category A ('blur') and category C ('motion blur'), and between category A ('JPEG compression') and category D ('no visible degradation'). A precise decision rule or examples of borderline cases would help reproducibility.
[Section II.A and III.A] The paper calls Qwen2.5-VL 'the leading open-source VLM' without citation or benchmark comparison. This is an unsupported superlative; it should be replaced by a specific reference or removed.
[Section IV] The experimental section would benefit from stating the model versions, hyperparameters, and any prompt parsing logic used to convert free-form VLM responses (e.g., 'A', 'Category B', or a sentence) into a routing decision. This information is essential for reproducibility.
[References] Several references are incomplete or inconsistently formatted: reference [4] lacks page numbers, reference [24] has a missing paper title and inconsistent author formatting, and the footnote in Section V contains a raw YouTube URL that should be replaced with a proper citation.

Circularity Check

0 steps flagged · score 1.0 of 10

No circular derivation; the pipeline is assembled from external pretrained models, and self-citations are not load-bearing.

full rationale

The manuscript contains no fitted parameters, no loss function, and no equation that maps an output back to an input. The pipeline is an assembly of externally pretrained models: Qwen2.5-VL for zero-shot classification, InvSR/Real-ESRGAN/PaddleOCR for super-resolution, YOLO/YOSO/LaMa/NAFNet for reflections, and NAFNet/CodeFormer for motion blur. Each component is cited from prior work, and the only novel decision is the hand-written routing prompt in Section III.A; that prompt is not derived from the restoration outputs. The paper's central assertion of "high accuracy in degradation classification and significant improvements in visual quality" (Sections I and IV) is not quantified in Section IV, which shows only three representative images per degradation type; this is a missing-evaluation problem, not a circularity. The authors do cite their own prior work ([16], [19]-[23], [27]), but those citations appear in related-work and super-resolution contexts and are not used to justify the VLM routing claim or to define the degradation categories. No self-citation chain forces the conclusion. Hence no circular step can be exhibited, and the score is minimal.

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

The paper introduces no new mathematical objects or fitted parameters. The pipeline is built entirely from pretrained models, and the central assumptions are about the reliability of VLM classification and the effectiveness of chained off-the-shelf models. These assumptions are untested because the experiments section contains only three example images with no quantitative evaluation.

assumptions (4)

domain assumption Qwen2.5-VL can accurately classify real-world degraded images into the four categories in a zero-shot manner using the provided prompt.
The entire pipeline routing depends on this assumption (Section III.A). No fine-tuning or evaluation is provided to support it.
domain assumption YOLO and YOSO can reliably detect strong reflections in arbitrary real-world images.
Reflection removal in Section III.C assumes these detection models produce accurate masks for LaMa inpainting.
domain assumption The specialized restoration models (InvSR, Real-ESRGAN, LaMa, NAFNet, CodeFormer) retain their standalone performance when chained in this pipeline.
The paper reports no end-to-end metrics, so it implicitly assumes that the integration does not degrade the individual models' performance.
domain assumption Routing to a specialized model is better than applying a single generic restoration model to all degradations.
This is the core motivation of the framework, but the paper provides no comparison against any single-model baseline.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Degradation-Aware Image Enhancement via Vision-Language Classification." pith.science (2026). https://pith.science/paper/L5ZF7EOI

@misc{pith2026250605450,
  author       = {Pith},
  title        = {Pith review of: Degradation-Aware Image Enhancement via Vision-Language Classification},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/L5ZF7EOI}},
  note         = {Machine review of arXiv:2506.05450}
}

read the original abstract

Image degradation is a prevalent issue in various real-world applications, affecting visual quality and downstream processing tasks. In this study, we propose a novel framework that employs a Vision-Language Model (VLM) to automatically classify degraded images into predefined categories. The VLM categorizes an input image into one of four degradation types: (A) super-resolution degradation (including noise, blur, and JPEG compression), (B) reflection artifacts, (C) motion blur, or (D) no visible degradation (high-quality image). Once classified, images assigned to categories A, B, or C undergo targeted restoration using dedicated models tailored for each specific degradation type. The final output is a restored image with improved visual quality. Experimental results demonstrate the effectiveness of our approach in accurately classifying image degradations and enhancing image quality through specialized restoration models. Our method presents a scalable and automated solution for real-world image enhancement tasks, leveraging the capabilities of VLMs in conjunction with state-of-the-art restoration techniques.

Figures

Figures reproduced from arXiv: 2506.05450 by the authors.

**Figure 2.** The architecture for InvSR super-resolution restoration, combined with text extraction and restoration for improved [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Reflection artifact removal architecture. It uses YOLO/YOSO for reflection detection and LaMa for inpainting the [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

Figures from the paper (2 more)

**Figure 4.** Figure 4: Motion blur restoration architecture, including NAFNet for deblurring and CodeFormer for facial enhancement. [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

**Figure 5.** Figure 5: Restoration results for different image degradation types. The three scenarios shown are: (1) Motion blur, (2) [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]

Discussion (0). Sign in 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

Dual-Domain Perspective on Degradation-Aware Fusion: A VLM-Guided Robust Infrared and Visible Image Fusion Framework
cs.CV 2025-09 conditional novelty 5.0 of 10

A VLM-prompt-guided network that jointly does wavelet-domain degradation suppression and spatial-domain fusion to merge degraded infrared and visible images end to end.

Reference graph

Works this paper leans on

27 extracted references · 25 canonical work pages · cited by 1 Pith paper

[1]

Arbitrary-steps image super-resolution via diffusion inversion,

Z. Yue, K. Liao, and C. C. Loy, “Arbitrary-steps image super-resolution via diffusion inversion,” CVPR, 2025

work page 2025
[2]

Paddleocr: An open-source op- tical character recognition system,

Baidu, “Paddleocr: An open-source op- tical character recognition system,” 2021, https://github.com/PaddlePaddle/PaddleOCR

work page 2021
[3]

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

X. Wang, L. Xie, C. Dong, and Y . Shan, “Real-esrgan: Training real-world blind super-resolution with pure synthetic data,” in CVPR, 2021, pp. 1905–1914

work page 2021
[4]

Yolov9: Learning what you want to learn using programmable gradient information,

C. Y . Wang, I. Yeh, and H.-Y . Mark Li, “Yolov9: Learning what you want to learn using programmable gradient information,” in ECCV. Springer, 2024

work page 2024
[5]

You only segment once: Towards real-time panoptic segmentation,

J. Hu, L. Huang, T. Ren, S. Zhang, R. Ji, and L. Cao, “You only segment once: Towards real-time panoptic segmentation,” in CVPR, 2023, pp. 17819–17829

work page 2023
[6]

Resolution-robust large mask inpainting with fourier convolutions,

R. Suvorov, E. Logacheva, A. Mashikhin, A. Remi- zova, A. Ashukha, A. Silvestrov, N. Kong, H. Goka, K. Park, and V . Lempitsky, “Resolution-robust large mask inpainting with fourier convolutions,” in CVPR, 2022, pp. 2149–2159

work page 2022
[7]

Simple baselines for image restoration,

L. Chen, X. Chu, X. Zhang, and J. Sun, “Simple baselines for image restoration,” in ECCV. Springer, 2022, pp. 17–33

work page 2022
[8]

Towards robust blind face restoration with codebook lookup transformer,

S. Zhou, K. Chan, C. Li, and C. C. Loy, “Towards robust blind face restoration with codebook lookup transformer,” NeurIPS, 2022

work page 2022

Show all 27 references

[9]

Visual instruction tuning,

H. Liu, C. Li, Q. Wu, and Y . J. Lee, “Visual instruction tuning,” NeurIPS, vol. 36, pp. 34892–34916, 2023

2023
[10]

Improved baselines with visual instruction tuning,

H. Liu, C. Li, Y . Li, and Y . J. Lee, “Improved baselines with visual instruction tuning,” in CVPR, 2024, pp. 26296–26306

2024
[11]

Blip: Bootstrap- ping language-image pre-training for unified vision- language understanding and generation,

J. Li, D. Li, C. Xiong, and S. Hoi, “Blip: Bootstrap- ping language-image pre-training for unified vision- language understanding and generation,” in ICML. PMLR, 2022, pp. 12888–12900

2022
[12]

Flamingo: a visual language model for few-shot learning,

J. B. Alayrac, J. Donahue, P. Luc, A. Miech, I. Barr, Y . Hasson, K. Lenc, A. Mensch, K. Millican, M. Reynolds, et al., “Flamingo: a visual language model for few-shot learning,” NeurIPS, vol. 35, pp. 23716–23736, 2022

2022
[13]

Qwen2.5- vl technical report,

S. Bai, K. Chen, X. Liu, J. Wang, W. Ge, S. Song, K. Dang, P. Wang, S. Wang, J. Tang, et al., “Qwen2.5- vl technical report,” arXiv preprint arXiv:2502.13923, 2025

2025 arXiv
[14]

Internvl3: Exploring advanced training and test-time recipes for open-source multimodal models,

J. Zhu, W. Wang, Z. Chen, Z. Liu, S. Ye, L. Gu, Y . Duan, H. Tian, W. Su, J. Shao, et al., “Internvl3: Exploring advanced training and test-time recipes for open-source multimodal models,” arXiv preprint arXiv:2504.10479, 2025

2025 arXiv
[15]

Learning transferable visual models from natural language supervision,

A. Radford, J. W. Kim, C. Hallacy, A. Ramesh, G. Goh, S. Agarwal, G. Sastry, A. Askell, P. Mishkin, J. Clark, et al., “Learning transferable visual models from natural language supervision,” in ICML, 2021

2021
[16]

Real- time super-resolution for real-world images on mobile devices,

J. Cai, Z. Meng, J. Ding, and C. M. Ho, “Real- time super-resolution for real-world images on mobile devices,” in MIPR. IEEE, 2022, pp. 127–132

2022
[17]

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

F. Yu, J. Gu, Z. Li, J. Hu, X. Kong, X. Wang, J. He, Y . Qiao, and C. Dong, “Scaling up to excellence: Practicing model scaling for photo-realistic image restoration in the wild,” in CVPR, 2024, pp. 25669– 25680

2024
[18]

One- step effective diffusion network for real-world image super-resolution,

R. Wu, L. Sun, Z. Ma, and L. Zhang, “One- step effective diffusion network for real-world image super-resolution,” NeurIPS, vol. 37, pp. 92529–92553, 2024

2024
[19]

Ntire 2025 challenge on single image reflection removal in the wild: Datasets, methods and results,

K. Yang, J. Cai, L. Ouyang, F. Vasluianu, R. Timofte, et al., “Ntire 2025 challenge on single image reflection removal in the wild: Datasets, methods and results,” in CVPR Workshops, 2025

2025
[20]

Openrr-1k: A scalable dataset for real-world reflection removal,

K. Yang, L. Ouyang, H. Sun, J. Cai, L. Fu, J. Ding, C. M. Ho, and Z. Meng, “Openrr-1k: A scalable dataset for real-world reflection removal,” in ICIP, 2025

2025
[21]

Survey on single-image reflection removal using deep learning techniques,

K. Yang, H. Sun, J. Cai, L. Fu, J. Ding, J. Li, and Z. Meng, “Survey on single-image reflection removal using deep learning techniques,” in MIPR, 2025

2025
[22]

F2t2-hit: A u-shaped fft transformer and hierarchical transformer for reflection removal,

J. Cai, K. Yang, L. Ouyang, L. Fu, J. Ding, H. Sun, C. M. Ho, and Z. Meng, “F2t2-hit: A u-shaped fft transformer and hierarchical transformer for reflection removal,” in ICIP, 2025

2025
[23]

Openrr-5k: A large-scale benchmark for reflection removal in the wild,

J. Cai, K. Yang, L. Ouyang, L. Fu, J. Ding, J. Shen, and Z. Meng, “Openrr-5k: A large-scale benchmark for reflection removal in the wild,” in MIPR, 2025

2025
[24]

Single image reflection sepa- ration via component synergy,

Q. Hu and X. Guo, “Single image reflection sepa- ration via component synergy,” in CVPR, 2023, pp. 13138–13147

2023
[25]

Revisiting single image reflection removal in the wild,

Y . Zhu, Xueyang Fu, Peng-Tao Jiang, Hao Zhang, Qibin Sun, Jinwei Chen, Zheng-Jun Zha, and Bo Li, “Revisiting single image reflection removal in the wild,” in CVPR, 2024, pp. 25468–25478

2024
[26]

Reversible de- coupling network for single image reflection removal,

H. Zhao, M. Li, Q. Hu, and X. Guo, “Reversible de- coupling network for single image reflection removal,” in CVPR, 2025

2025
[27]

Joint hdr denoising and fusion on mobile devices,

J. Cai, Y . Lin, J. Li, J. Ding, L. Ouyang, C. M. Ho, and Z. Meng, “Joint hdr denoising and fusion on mobile devices,” in MIPR. IEEE, 2024, pp. 247–252

2024

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

Arbitrary-steps image super-resolution via diffusion inversion,

Z. Yue, K. Liao, and C. C. Loy, “Arbitrary-steps image super-resolution via diffusion inversion,” CVPR, 2025

work page 2025

[2] [2]

Paddleocr: An open-source op- tical character recognition system,

Baidu, “Paddleocr: An open-source op- tical character recognition system,” 2021, https://github.com/PaddlePaddle/PaddleOCR

work page 2021

[3] [3]

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

X. Wang, L. Xie, C. Dong, and Y . Shan, “Real-esrgan: Training real-world blind super-resolution with pure synthetic data,” in CVPR, 2021, pp. 1905–1914

work page 2021

[4] [4]

Yolov9: Learning what you want to learn using programmable gradient information,

C. Y . Wang, I. Yeh, and H.-Y . Mark Li, “Yolov9: Learning what you want to learn using programmable gradient information,” in ECCV. Springer, 2024

work page 2024

[5] [5]

You only segment once: Towards real-time panoptic segmentation,

J. Hu, L. Huang, T. Ren, S. Zhang, R. Ji, and L. Cao, “You only segment once: Towards real-time panoptic segmentation,” in CVPR, 2023, pp. 17819–17829

work page 2023

[6] [6]

Resolution-robust large mask inpainting with fourier convolutions,

R. Suvorov, E. Logacheva, A. Mashikhin, A. Remi- zova, A. Ashukha, A. Silvestrov, N. Kong, H. Goka, K. Park, and V . Lempitsky, “Resolution-robust large mask inpainting with fourier convolutions,” in CVPR, 2022, pp. 2149–2159

work page 2022

[7] [7]

Simple baselines for image restoration,

L. Chen, X. Chu, X. Zhang, and J. Sun, “Simple baselines for image restoration,” in ECCV. Springer, 2022, pp. 17–33

work page 2022

[8] [8]

Towards robust blind face restoration with codebook lookup transformer,

S. Zhou, K. Chan, C. Li, and C. C. Loy, “Towards robust blind face restoration with codebook lookup transformer,” NeurIPS, 2022

work page 2022