Pith. sign in

REVIEW 3 major objections 5 minor 45 references

Predicting the Original Appearance of Damaged Historical Documents

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read The paper claims that a diffusion-based network, DiffHDR, can predict the original appearance of damaged historical documents when given the damaged image plus a content image and mask of the missing characters, and that it outperforms…

desk verdict A genuinely useful benchmark dataset and a strong conditional inpainter, but the title and abstract promise autonomous reconstruction the method doesn't deliver. read the letter →

arxiv 2412.11634 v1 pith:MZQFXBSS submitted 2024-12-16 cs.CV

Zhenhua Yang , Dezhi Peng , Yongxin Shi , Yuyi Zhang , Chongyu Liu , Lianwen Jin This is my paper · ORCID
classification cs.CV
keywords historicaldocumentrepairdiffusionmodelsimagerestorationHDR28Kdatasetcharacterperceptuallossculturalheritagedigitizationinpaintingclassifier-freeguidance
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

This paper introduces Historical Document Repair as a new task and argues that the original appearance of a damaged historical document can be predicted, not just binarized or enhanced. It contributes HDR28K, a dataset of 28,552 damaged-repaired image pairs built from two Chinese historical document collections with three synthetic degradations: character missing, paper damage, and ink erosion. The proposed DiffHDR is a diffusion model that takes the damaged patch together with a user-supplied content image and damage mask as conditions, and directly produces the repaired image. On the HDR28K test set, the authors report that DiffHDR outperforms nine existing restoration and generation methods on FID, LPIPS, and repaired-character recognition accuracy, and they show qualitative repairs of real damaged scans. The broader promise is a practical tool for cultural-heritage conservation that can reconstruct missing text in context.

What carries the argument

The load-bearing mechanism is an 8-channel conditional diffusion denoiser that concatenates the noised image, the damaged image, the content image, and the mask image, and outputs the repaired image in a single forward pass. The content image supplies the semantics of what should appear and the mask supplies the spatial location, so the model does not need to infer damaged content from pixels alone. Randomly dropping these conditions during training enables classifier-free guidance at inference, with separate guidance scales for the damaged image and for the semantic/spatial pair. The character perceptual loss, computed as a masked multi-scale VGG feature distance between the repaired and target regions, focuses training on producing correct character content and style within the repaired area.

What would settle it

Feed DiffHDR real damaged document patches with severe ink erosion or large missing regions while suppressing the content and mask cues (as in the paper's limitation figure) and measure the repaired-region character recognition accuracy; the paper's own limitation predicts it will remain near chance. A complementary test is to supply a deliberately wrong transcript for one damaged character: if the model faithfully renders the wrong character, that confirms the repair is copying the supplied cue rather than inferring the original content, which is exactly the conditional behavior the paper claims.

Watch

Extended reading notes

Core claim

The central claim is that a conditional diffusion process can reconstruct the original appearance of damaged historical documents when it is told what the missing characters say and where they sit. Formally, the model learns $x_r = F(x_d, x_c, x_m)$, where $x_d$ is the damaged image, $x_c$ is a content image encoding the transcript of the damaged characters, and $x_m$ is a mask marking the damaged regions. DiffHDR's denoiser predicts the repaired image directly, rather than the added noise, and is trained with a diffusion loss plus a character perceptual loss that penalizes feature mismatches inside the repaired regions. The authors report that this design outperforms the compared methods on the HDR28K benchmark and that the model trained only on synthetic degradations still repairs real damaged documents, edits text content, and generates new text blocks in the document's style.

Load-bearing premise

At test time, someone must already know and supply the content and exact location of every damaged character, and if the damage is severe the method cannot repair the image without those cues.

Editorial extensions

If this is right

  • HDR28K gives future work a common benchmark of 28,552 damaged-repaired pairs with character-level annotations, so new repair methods can be trained and compared on the same ground.
  • A model trained only on synthetic degradations transfers to real damaged scans, which means large-scale synthetic data can substitute for scarce real damaged-repaired pairs.
  • The same conditional diffusion framework performs document editing and text block font generation, so one model can rewrite or extend a historical document in its own style.
  • Repaired-region character recognition accuracy is substantially higher with DiffHDR than with the compared baselines, implying fewer hallucinated or wrong characters in the output.

Reading between the lines

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

  • The method's reliance on user-provided transcripts means an end-to-end repair tool would need an upstream module that reads partially damaged text; the paper names this as future work but does not implement it.
  • Because HDR28K is built from Chinese historical documents and three synthetic damage types, transfer to other scripts, inks, and damage modes is untested; analogous datasets for other traditions would be a direct extension.
  • The conditional formulation suggests a controlled-use application: by deliberately altering the content image, the same model could serve as a text-editing or reconstruction tool for digital heritage, a capability the paper only begins to explore.
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.

Referee Report

3 major / 5 minor

Summary. The paper introduces Historical Document Repair (HDR), a task of reconstructing the original appearance of damaged historical document images. It contributes HDR28K, a dataset of 28,552 synthetic damaged-repaired pairs with character-level annotations, and DiffHDR, a diffusion-based model conditioned on the damaged image xd, a content image xc, and a mask image xm, together with a character perceptual loss. Experiments on HDR28K compare against nine baselines and report lower FID/LPIPS and higher character-recognition accuracy, along with qualitative results on real damaged documents and extensions to document editing and text-block generation.

Significance. The dataset and the guided-repair formulation are potentially useful contributions: HDR28K is large, multi-style, and accompanied by character-level annotations; the held-out evaluation is reasonably designed with no same-document leakage between train and test; the comparison is broad; and the announced code and data release would support reproducibility. The reported gains on the guided task are substantial. However, the significance as currently framed is overstated: the method does not predict unknown damaged content but renders user-specified content into user-specified locations. If repositioned as an interactive repair tool, the work is solid; the autonomous 'predicting the original appearance' claim requires an unconditional model or a demonstration without user-supplied cues.

major comments (3)
  1. [Abstract; Eq. (2); Section 'DiffHDR: Diffusion-based HDR Network' (Condition Parsing); Limitation] The central claim of predicting the original appearance of damaged documents is not supported for the autonomous setting. The task is defined in Eq. (1) as xr = FHDR(xd), but the proposed method in Eq. (2) is xr = F(xd, xc, xm), where xc and xm are parsed from user-provided content and location. The condition-parsing paragraph explicitly states that 'the user provides the content and location of the damaged characters,' and the Limitation section concedes that without semantic and spatial information, severe damage cannot be repaired. The abstract and title should be qualified, or the paper should report results without xc and xm and quantify the gap.
  2. [Evaluation Metrics; Table 1] Rec-ACC is measured with xc supplied, because the recognizer is applied to repaired regions that were generated from the ground-truth content image. This metric therefore measures how faithfully the model renders given content, not how well it recovers unknown content. The 'character correctness' claim in Table 1 should be stated in conditional terms; an unconditional evaluation (for example, recognition accuracy on outputs produced without xc) is needed to support the autonomous framing.
  3. [Real Damaged Document Image Repair (Figure 9)] The real-document experiment does not report what inputs were provided: whether xc and xm were supplied and, if so, whether they came from manual transcription or an automated recognizer. Without this information, the qualitative results support only guided repair, not autonomous damage recovery. Please state the protocol and, if feasible, include a failure analysis when the cues are withheld.
minor comments (5)
  1. [Related Work heading] The heading 'Historcial Document Image Processing' contains a typo; it should be 'Historical.'
  2. [Eq. (2) explanation] The word 'senmantic' in the explanation of xc should be 'semantic.'
  3. [Eq. (6)] The notation 'sc.m' should be 'sc,m' to match the text describing the guidance scales.
  4. [Figure 10 caption] The caption 'without semantic and visual cues' is ambiguous because the first row shows successful repair; the text clarifies this, but the figure should distinguish the moderate case from the severe failure case.
  5. [Attribute-Sensitive Repair Strategy] The sentence describing the 8% dropout is ambiguous about the probabilities for the three conditions; please specify the exact sampling schedule (for example, each condition with 8% independently or jointly).

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the model is a supervised conditional repair network whose user-provided content and mask inputs are disclosed, the evaluation is held-out and compares against baselines receiving the same cues, and the limitation that severe damage cannot be repaired without cues is explicitly acknowledged.

full rationale

The derivation chain is self-contained and does not reduce to its inputs. The paper defines the HDR task in Eq. (1) as xr = FHDR(xd), but the proposed method in Eq. (2) explicitly conditions on user-supplied semantic and spatial cues: xr = F(xd, xc, xm). The condition-parsing section states 'the user provides the content and location of the damaged characters and we parse them out to obtain the content image xc and the mask image xm.' This means the system is a guided inpainter rather than an autonomous content predictor, but this is disclosed in the method and reiterated in the Limitation section ('when the damage is severe, our method is unable to repair the image in the absence of semantic and spatial information'). The held-out test set is constructed from source-dataset test images with no patch overlap with training, and all compared methods are given the same concatenated inputs xd, xc, xm, so the quantitative gains are not forced by giving DiffHDR extra information. Rec-ACC is measured with xc supplied, so it evaluates rendering fidelity to given content rather than recovery of unknown content; that is a scope limitation, not a circular derivation. The citations to prior same-group works (MTHv2, M5HisDoc, UPOCR, FontDiffuser) are used as dataset sources, baselines, or related work, not as load-bearing unverified theorems, and no uniqueness claim is imported from the authors. The title and abstract phrase 'predict the original appearance' is broader than the implemented conditional repair, but the paper itself acknowledges the narrower capability, so the gap is a correctness or scope concern, not circularity.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The paper introduces no new physical entities. The main contributions are a dataset and a conditional diffusion model. The free parameters are limited to guidance scales and training hyperparameters. The key domain assumptions are about the realism of synthetic degradations and the availability of content and mask cues.

free parameters (3)
  • Guidance scale s_d = 1.2
    Classifier-free guidance weight for the damaged image condition, chosen by hand in the Implementation Details section.
  • Guidance scale s_c,m = 1.5
    Classifier-free guidance weight for the content and mask conditions, chosen by hand in the Implementation Details section.
  • Conditional dropout probability = 8%
    Hyperparameter for training with classifier-free guidance, chosen in the Implementation Details section.
assumptions (4)
  • domain assumption Synthetic degradations (character missing via LaMa, paper damage masks, ink erosion via genalog) faithfully simulate real historical document damage.
    The dataset construction section describes the degradations; the real-document generalization claim rests on this transfer. Only qualitative evidence supports it.
  • domain assumption MTHv2 and M5HisDoc character annotations provide accurate ground-truth content for all damaged regions.
    The dataset is built from these sources; any annotation errors propagate to the content image x_c and to the Rec-ACC evaluation.
  • domain assumption Pretrained VGG features are a suitable perceptual space for character content preservation.
    The character perceptual loss in Eq. (5) uses VGG19; this is a standard assumption in image generation.
  • standard math The standard DDPM forward process equations hold.
    Used in Eq. (3) to define the noised image x_t and the noise schedule.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Predicting the Original Appearance of Damaged Historical Documents." pith.science (2026). https://pith.science/paper/MZQFXBSS

@misc{pith2026241211634,
  author       = {Pith},
  title        = {Pith review of: Predicting the Original Appearance of Damaged Historical Documents},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MZQFXBSS}},
  note         = {Machine review of arXiv:2412.11634}
}
read the original abstract

Historical documents encompass a wealth of cultural treasures but suffer from severe damages including character missing, paper damage, and ink erosion over time. However, existing document processing methods primarily focus on binarization, enhancement, etc., neglecting the repair of these damages. To this end, we present a new task, termed Historical Document Repair (HDR), which aims to predict the original appearance of damaged historical documents. To fill the gap in this field, we propose a large-scale dataset HDR28K and a diffusion-based network DiffHDR for historical document repair. Specifically, HDR28K contains 28,552 damaged-repaired image pairs with character-level annotations and multi-style degradations. Moreover, DiffHDR augments the vanilla diffusion framework with semantic and spatial information and a meticulously designed character perceptual loss for contextual and visual coherence. Experimental results demonstrate that the proposed DiffHDR trained using HDR28K significantly surpasses existing approaches and exhibits remarkable performance in handling real damaged documents. Notably, DiffHDR can also be extended to document editing and text block generation, showcasing its high flexibility and generalization capacity. We believe this study could pioneer a new direction of document processing and contribute to the inheritance of invaluable cultures and civilizations. The dataset and code is available at https://github.com/yeungchenwa/HDR.

Figures

Figures reproduced from arXiv: 2412.11634 by the authors.

Figure 1
Figure 1. Definition of Historical Document Repair (HDR) [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Damaged Samples in HDR28K. et al. 2022) are primarily focused on image deraining, de￾fogging, deblurring, or denoising. For example, Restormer (Zamir et al. 2022) proposes an efficient transformer model to capture long-range pixel interactions and is applicable to large images. In document image restoration, they mainly address the tasks of rectification (Li et al. 2023a; Jiang et al. 2022; Das et al. 2019), binariz… view at source ↗
Figure 3
Figure 3. Construction pipeline of the HDR28K dataset. [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (13 more)
Figure 4
Figure 4. Figure 4: Statistics of the HDR28K dataset. Data Construction To accurately simulate real damaged scenarios in historical documents, it is necessary to degrade the characters, which requires character-level bounding boxes and content anno￾tations. Therefore, we constructed the H…
Figure 5
Figure 5. Figure 5: Overview of our proposed method. DiffHDR comprises a condition parsing and a diffusion pipeline. In the condition [PITH_FULL_IMAGE:figures/full_fig_p004_5.png]
Figure 7
Figure 7. Figure 7: Document editing and text block font generation. [PITH_FULL_IMAGE:figures/full_fig_p005_7.png]
Figure 6
Figure 6. Figure 6: Unsuitableness of PSNR and SSIM. results with the damaged image xd and the content and loca￾tion cues xc, xm, respectively. Thus, the repair strategy can be formulated as: F˜(xt;xd, xc, xm) = F(xt; ∅, ∅, ∅) + sd(F(xt; xd, ∅, ∅) − F(xt; ∅, ∅, ∅)) + sc.m(F(xt; xd, xc, xm…
Figure 8
Figure 8. Figure 8: Qualitative comparison. We visualize the results of some evaluated methods. The green, blue, and red boxes represent [PITH_FULL_IMAGE:figures/full_fig_p006_8.png]
Figure 9
Figure 9. Figure 9: Real damaged historical documents repair by DiffHDR. [PITH_FULL_IMAGE:figures/full_fig_p007_9.png]
Figure 10
Figure 10. Figure 10: Damaged historical documents repair by DiffHDR when not provided with semantic and spatial cues. Editing and Text Block Font Generation In this section, we explore the capabilities of DiffHDR in historical document editing and text block font generation. (1) Document …
Figure 11
Figure 11. Figure 11: Examples of diverse backgrounds in HDR28K. [PITH_FULL_IMAGE:figures/full_fig_p010_11.png]
Figure 12
Figure 12. Figure 12: Examples of diverse text densities in HDR28K. [PITH_FULL_IMAGE:figures/full_fig_p010_12.png]
Figure 14
Figure 14. Figure 14: Examples of diverse character styles in HDR28K. [PITH_FULL_IMAGE:figures/full_fig_p010_14.png]
Figure 15
Figure 15. Figure 15: Historical document repair on the type of character missing in HDR28K [PITH_FULL_IMAGE:figures/full_fig_p011_15.png]
Figure 16
Figure 16. Figure 16: Historical document repair on the type of paper damage in HDR28K [PITH_FULL_IMAGE:figures/full_fig_p011_16.png]
Figure 17
Figure 17. Figure 17: Historical document repair on the type of ink erosion in HDR28K [PITH_FULL_IMAGE:figures/full_fig_p012_17.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

45 extracted references · 28 canonical work pages

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    Amin, J.; Siddiqi, I.; and Moetesum, M. 2023. Reconstruction of Broken Writing Strokes in Greek Papyri. In International Conference on Document Analysis and Recognition, 253--266. Springer

    work page 2023

  4. [4]

    Assael, Y.; Sommerschield, T.; Shillingford, B.; Bordbar, M.; Pavlopoulos, J.; Chatzipanagiotou, M.; Androutsopoulos, I.; Prag, J.; and de Freitas, N. 2022. Restoring and attributing ancient texts using deep neural networks. Nature, 603(7900): 280--283

    work page 2022

  5. [5]

    Brooks, T.; Holynski, A.; and Efros, A. A. 2023. Instructpix2pix: Learning to follow image editing instructions. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 18392--18402

    2023

  6. [6]

    Chen, L.; Chu, X.; Zhang, X.; and Sun, J. 2022. Simple baselines for image restoration. In European Conference on Computer Vision, 17--33. Springer

    work page 2022

  7. [7]

    Chen, Z.; Wu, J.; Wang, W.; Su, W.; Chen, G.; Xing, S.; Zhong, M.; Zhang, Q.; Zhu, X.; Lu, L.; et al. 2024. InternVL : Scaling up vision foundation models and aligning for generic visual-linguistic tasks. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 24185--24198

    work page 2024

  8. [8]

    Cui, Y.; Ren, W.; Cao, X.; and Knoll, A. 2023. Focal Network for Image Restoration. In Proceedings of the IEEE/CVF international conference on computer vision, 13001--13011

    work page 2023

Show all 45 references
  1. [9]

    Das, S.; Ma, K.; Shu, Z.; Samaras, D.; and Shilkrot, R. 2019. Dewarpnet: Single-image document unwarping with stacked 3d and 2d regression networks. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 131--140

    2019

  2. [10]

    E.-A.; and Cheriet, M

    Ech-Cherif, M. E.-A.; and Cheriet, M. 2022. Frank-Wolfe-based multi-task learning for historical document restoration. In 2022 26th International Conference on Pattern Recognition (ICPR), 3900--3907. IEEE

    2022

  3. [11]

    Hedjam, R.; and Cheriet, M. 2013. Historical document image restoration using multispectral imaging system. Pattern Recognition, 46(8): 2297--2312

    2013

  4. [12]

    Hertlein, F.; and Naumann, A. 2023. Template-Guided Illumination Correction for Document Images with Imperfect Geometric Reconstruction. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 904--913

    2023

  5. [13]

    Heusel, M.; Ramsauer, H.; Unterthiner, T.; Nessler, B.; and Hochreiter, S. 2017. Gans trained by a two time-scale update rule converge to a local nash equilibrium. Advances in neural information processing systems, 30

    2017

  6. [14]

    Ho, J.; Jain, A.; and Abbeel, P. 2020. Denoising diffusion probabilistic models. Advances in neural information processing systems, 33: 6840--6851

    2020

  7. [15]

    Ho, J.; and Salimans, T. 2022. Classifier-free diffusion guidance. arXiv preprint arXiv:2207.12598

    2022 arXiv

  8. [16]

    Isola, P.; Zhu, J.-Y.; Zhou, T.; and Efros, A. A. 2017. Image-to-image translation with conditional adversarial networks. In Proceedings of the IEEE conference on computer vision and pattern recognition, 1125--1134

    2017

  9. [17]

    Jiang, X.; Long, R.; Xue, N.; Yang, Z.; Yao, C.; and Xia, G.-S. 2022. Revisiting document image dewarping by grid regularization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 4543--4552

    2022

  10. [18]

    Kong, Y.; Luo, C.; Ma, W.; Zhu, Q.; Zhu, S.; Yuan, N.; and Jin, L. 2022. Look closer to supervise better: one-shot font generation via component-based discriminator. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 13482--13491

    2022

  11. [19]

    Li, H.; Wu, X.; Chen, Q.; and Xiang, Q. 2023 a . Foreground and Text-lines Aware Document Image Rectification. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 19574--19583

    2023

  12. [20]

    Li, Y.; Fan, Y.; Xiang, X.; Demandolx, D.; Ranjan, R.; Timofte, R.; and Van Gool, L. 2023 b . Efficient and explicit modelling of image hierarchies for image restoration. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 18278--18289

    2023

  13. [21]

    Li, Z.; Chen, X.; Pun, C.-M.; and Cun, X. 2023 c . High-Resolution Document Shadow Removal via A Large-Scale Real-World Dataset and A Frequency-Aware Shadow Erasing Net. arXiv preprint arXiv:2308.14221

    2023 arXiv

  14. [22]

    Lin, Y.-H.; Chen, W.-C.; and Chuang, Y.-Y. 2020. Bedsr-net: A deep shadow removal network from a single document image. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 12905--12914

    2020

  15. [23]

    Lu, C.; Zhou, Y.; Bao, F.; Chen, J.; Li, C.; and Zhu, J. 2022. Dpm-solver++: Fast solver for guided sampling of diffusion probabilistic models. arXiv preprint arXiv:2211.01095

    2022 arXiv

  16. [24]

    Ma, W.; Zhang, H.; Jin, L.; Wu, S.; Wang, J.; and Wang, Y. 2020. Joint layout analysis, character detection and recognition for historical document digitization. In 2020 17th International Conference on Frontiers in Handwriting Recognition (ICFHR), 31--36. IEEE

    2020

  17. [25]

    C.; Nguyen, C

    Nguyen, K. C.; Nguyen, C. T.; Hotta, S.; and Nakagawa, M. 2019. A character attention generative adversarial network for degraded historical document restoration. In 2019 International Conference on Document Analysis and Recognition (ICDAR), 420--425. IEEE

    2019

  18. [26]

    Peng, D.; Yang, Z.; Zhang, J.; Liu, C.; Shi, Y.; Ding, K.; Guo, F.; and Jin, L. 2023. UPOCR: Towards unified pixel-level ocr interface. In Forty-first International Conference on Machine Learning

    2023

  19. [27]

    Raha, P.; and Chanda, B. 2019. Restoration of historical document images using convolutional neural networks. In 2019 IEEE region 10 symposium (TENSYMP), 56--61. IEEE

    2019

  20. [28]

    Ronneberger, O.; Fischer, P.; and Brox, T. 2015. U-net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention--MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part II...

    2015

  21. [29]

    Shi, Y.; Liu, C.; Peng, D.; Jian, C.; Huang, J.; and Jin, L. 2023. M5HisDoc: A Large-scale Multi-style Chinese Historical Document Analysis Benchmark. In Thirty-seventh Conference on Neural Information Processing Systems Datasets and Benchmarks Track

    2023

  22. [30]

    Simonyan, K.; and Zisserman, A. 2014. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556

    2014 arXiv

  23. [31]

    Suvorov, R.; Logacheva, E.; Mashikhin, A.; Remizova, A.; Ashukha, A.; Silvestrov, A.; Kong, N.; Goka, H.; Park, K.; and Lempitsky, V. 2022. Resolution-robust large mask inpainting with fourier convolutions. In Proceedings of the IEEE/CVF winter conference on applications of co...

    2022

  24. [32]

    Wadhwani, M.; Kundu, D.; Chakraborty, D.; and Chanda, B. 2021. Text extraction and restoration of old handwritten documents. Digital Techniques for Heritage Presentation and Preservation, 109--132

    2021

  25. [33]

    Wang, C.; Zhou, M.; Ge, T.; Jiang, Y.; Bao, H.; and Xu, W. 2023. CF-Font: Content Fusion for Few-shot Font Generation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1858--1867

    2023

  26. [34]

    Wang, P.; Bai, S.; Tan, S.; Wang, S.; Fan, Z.; Bai, J.; Chen, K.; Liu, X.; Wang, J.; Ge, W.; et al. 2024. Qwen2-VL : Enhancing vision-language model's perception of the world at any resolution. arXiv preprint arXiv:2409.12191

    2024 arXiv

  27. [35]

    Wang, Y.; Zhou, W.; Lu, Z.; and Li, H. 2022 a . Udoc-gan: Unpaired document illumination correction with background light prior. In Proceedings of the 30th ACM International Conference on Multimedia, 5074--5082

    2022

  28. [36]

    Wang, Z.; Cun, X.; Bao, J.; Zhou, W.; Liu, J.; and Li, H. 2022 b . Uformer: A general u-shaped transformer for image restoration. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 17683--17693

    2022

  29. [37]

    Xue, C.; Tian, Z.; Zhan, F.; Lu, S.; and Bai, S. 2022. Fourier document restoration for robust document dewarping and recognition. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 4573--4582

    2022

  30. [38]

    Yang, M.; and Xu, S. 2023. A novel Degraded Document Binarization model through vision transformer network. Information Fusion, 93: 159--173

    2023

  31. [39]

    Yang, Z.; Liu, B.; Xxiong, Y.; Yi, L.; Wu, G.; Tang, X.; Liu, Z.; Zhou, J.; and Zhang, X. 2023 a . DocDiff: Document enhancement via residual diffusion models. In Proceedings of the 31st ACM International Conference on Multimedia, 2795--2806

    2023

  32. [40]

    Yang, Z.; Peng, D.; Kong, Y.; Zhang, Y.; Yao, C.; and Jin, L. 2023 b . FontDiffuser: One-Shot Font Generation via Denoising Diffusion with Multi-Scale Content Aggregation and Style Contrastive Learning. arXiv preprint arXiv:2312.12142

    2023 arXiv

  33. [41]

    W.; Arora, A.; Khan, S.; Hayat, M.; Khan, F

    Zamir, S. W.; Arora, A.; Khan, S.; Hayat, M.; Khan, F. S.; and Yang, M.-H. 2022. Restormer: Efficient transformer for high-resolution image restoration. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 5728--5739

    2022

  34. [42]

    M.; and Shum, H.-Y

    Zhang, H.; Li, F.; Liu, S.; Zhang, L.; Su, H.; Zhu, J.; Ni, L. M.; and Shum, H.-Y. 2022. DINO : Detr with improved denoising anchor boxes for end-to-end object detection. arXiv preprint arXiv:2203.03605

    2022 arXiv

  35. [43]

    A.; Shechtman, E.; and Wang, O

    Zhang, R.; Isola, P.; Efros, A. A.; Shechtman, E.; and Wang, O. 2018. The unreasonable effectiveness of deep features as a perceptual metric. In Proceedings of the IEEE conference on computer vision and pattern recognition, 586--595

    2018

  36. [44]

    Zhu, J.-Y.; Park, T.; Isola, P.; and Efros, A. A. 2017. Unpaired image-to-image translation using cycle-consistent adversarial networks. In Proceedings of the IEEE international conference on computer vision, 2223--2232

    2017

  37. [45]

    Zhu, S.; Xue, H.; Nie, N.; Zhu, C.; Liu, H.; and Fang, P. 2024. Reproducing the Past: A Dataset for Benchmarking Inscription Restoration. In Proceedings of the 32nd ACM International Conference on Multimedia, 7714--7723

    2024

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.