Open-Set Visual Text Forensics via Sparse-Constraint Rectified Flow

Jiangling Zhang; Shuxuan Gao; Yichao Liu; Yu Zhou; Zeyu Chen

REVIEW 3 major objections 5 minor 49 references

Open-Set Visual Text Forensics via Sparse-Constraint Rectified Flow

T0 review · 3 major / 5 minor · reviewed 2026-08-04 · deepseek-v4-flash

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

Pith's one-line read Text tampering can be localized by measuring, per pixel, how much a learned flow would need to move that patch to restore authentic text statistics — no forgery-specific training needed.

desk verdict Solid flow-matching approach to text tampering localization with good ablations, but the zero-shot claim is ahead of the evidence: the detector is trained on five hand-picked artifact types, and nothing shows that real text-edits stay inside that span. read the letter →

arxiv 2608.02258 v1 pith:MX57FBC2 submitted 2026-08-03 cs.CV cs.AI

Jiangling Zhang , Shuxuan Gao , Zeyu Chen , Yichao Liu , Yu Zhou This is my paper · ORCID

classification cs.CVcs.AI

keywords Open-setvisualtextforensicstamperingdetectionFlowmatchingSparseanomalylocalizationSelf-supervisedartifactinjectionPixel-spacediffusiontransformerRestorationcostZero-shotgeneralization

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 tries to establish that open-set text forgery detection is better cast as measuring local deviation from authentic visual-text statistics than as classifying known forgery patterns. To do this, it trains a generative model — Sparse-Constraint Rectified Flow — to predict the velocity field that would restore a locally corrupted image to an authentic one; the magnitude of that velocity at each pixel, evaluated at time zero, becomes the tampering score. Because tampering is spatially sparse, the flow objective is reweighted by the tampering mask so the learned field concentrates on anomalous regions instead of collapsing to an identity map on the background. Training pairs come from a self-supervised artifact-injection scheme applied to authentic text images, and the architecture operates in pixel space to preserve high-frequency forensic traces. The authors report the highest F1 and IoU among compared methods on three benchmarks, with particular strength in zero-shot transfer to unseen editing patterns.

What carries the argument

Sparse-Constraint Rectified Flow (SC-RF) — a spatially weighted flow-matching objective that regresses a velocity field along the straight-line transport from a locally corrupted image to its authentic counterpart. The spatial weight map 1 + λ·M_s reweights the loss toward tampered pixels, preventing the 'lazy identity' solution that standard flow matching produces on sparse anomalies. The detector uses the magnitude of the predicted velocity at t=0 as the tampering score, and the architecture estimating it, Forensic-DiT, operates in pixel space (no VAE compression) while fusing RGB, SRM noise residuals, and block-DCT frequency tokens; it also injects an image-specific forensic fingerprint v

What would settle it

Run the zero-shot model on text tampered by a newly deployed editor whose artifacts deliberately fall outside the injection span (e.g., sub-pixel warping, GAN-painting without blending mismatch, or compression-aware editing that exactly matches the background noise). If the F1 score drops to chance levels on such edits while staying high on in-span edits, the open-set generalization claim fails for that artifact class. A complementary test: apply the model to authentic images with benign local variation like glossy highlights or shadows, and check whether the restoration field also fires on th

Watch

Extended reading notes

Core claim

The central discovery is that a detector-oriented flow-matching objective can turn the generative task of restoring corrupted text into a pixel-level anomaly score. The paper redefines the source and target distributions of rectified flow: instead of Gaussian noise, the source is a locally corrupted version of an authentic image, and the target is the authentic image itself. The learned conditional velocity field is sparse — nonzero only where the image was corrupted — so SC-RF reweights the regression loss by the tampering mask, with weight W_s = 1 + λ·M_s. At inference, no ODE integration is needed: the L2 norm of the predicted velocity at t=0 is the local restoration cost, and this score

Load-bearing premise

The detector is trained only on synthetic artifacts — blur, JPEG compression, noise, alpha-blending mismatch, and glyph re-rendering — and the method assumes these injections produce restoration residuals representative of real, unseen text edits; if a real editor leaves traces outside this span, the learned field will not fire.

Editorial extensions

If this is right

Zero-shot localization generalizes across editing pipelines because the detector never sees specific forgery signatures during training.
One-step inference suffices at test time, so the method is practical: no ODE integration is required beyond predicting the velocity field at t=0.
The training paradigm needs only authentic images plus synthetic artifact injection, so it can extend to new domains without collecting tampered examples.
The stress-test result implies the same architecture can serve as an adversarial audit tool: local harmonization weakens the cues that existing forensic detectors rely on.

Reading between the lines

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

The synthetic-to-real boundary is the natural failure horizon: if future text editors produce traces outside the artifact-injection span (e.g., sub-pixel warping, GAN painting without blending mismatch, noise-matched inpainting), the restoration-cost field should be expected to miss them — a testable prediction that follows from the training setup rather than a claim made in the paper.
Because the detector never conditions on text semantics, the same restoration-cost formulation is a plausible transfer to other sparse tampering domains such as faces, objects, or document layout, which the paper does not evaluate.
The partial harmonization result suggests the learned velocity direction encodes an interpretable axis pointing from tampered to authentic; aligning that axis with specific artifact classes (blur vs. noise vs. compression) could yield explainable detections.

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.

Desk Editor's Note

read the letter

This is a well-executed, mostly convincing paper on open-set text tampering localization, but the zero-shot generalization claim is softer than the abstract suggests. The core idea - learn a velocity field that restores locally corrupted text regions, then use its magnitude at t=0 as an anomaly score - is sensible, and the sparse-constraint reweighting is a simple fix for a real problem: standard flow matching ignores sparse anomalies. The Forensic-DiT with SRM and DCT tokenization and the pixel-space design are reasonable engineering choices, and the ablations show each component contributes. The comparison against direct residual regression under the same backbone is a clean experiment; it gives the method credit for the objective, not just the architecture. I also appreciate that the authors are explicit that their authentic distribution is tied to the training corpus, not a universal manifold. That is honest. The soft spots are mostly around the zero-shot evaluation. First, the Artifact Injection uses five corruption types: blur, JPEG, noise, alpha-blending mismatch, and glyph re-rendering. The OSTF benchmark is supposed to show generalization to unseen text editing patterns, but there is no analysis of what those editing pipelines actually produce, or whether they fall within the span of those five operations. If real tampering involves diffusion-based inpainting or seamless copy-move, the detector may simply not have learned to fire on those statistics. The paper does not test that, so the strong zero-shot performance is really strong performance on a benchmark whose edits happen to resemble the injected artifacts. That is a testable claim, and it should be tested. Second, the F1/IoU thresholds are selected per benchmark on the validation set, which is standard practice but means the headline numbers are partly fitted to each evaluation set. Minor, but worth stating. Third, the training corpus includes ICDAR2013, from which T-IC13 is derived; the authors follow the DAF protocol, so this is not a hidden leak, but it does weaken the T-IC13 zero-shot result as a true held-out test. I would call this a moderate concern, not a fatal one. The math is sound - the objective is a weighted L2 regression with a correct target - and the citation pattern is fine. The paper deserves a serious referee. I would suggest the authors be asked to either broaden the artifact injection set or explicitly measure the spectral/residual overlap between the injected corruptions and the edits in OSTF, and to report threshold-free PR curves or error bars on the zero-shot numbers. As it stands, it is a good paper with a slightly overreaching conclusion. Send it to peer review.

Referee Report

3 major / 5 minor

Summary. The paper proposes SC-RF, a flow-matching-based detector for open-set visual text tampering. The method trains a pixel-space Forensic-DiT to regress the residual between a locally corrupted image and its authentic counterpart, using a spatially weighted rectified-flow objective (Eq. 7). At inference, the L2 norm of the predicted velocity at t=0 is used as a pixel-level tampering score. Training is self-supervised via Artifact Injection, which applies five local corruption types to authentic scene-text and document images. Experiments on T-IC13, DocTamper, and OSTF report state-of-the-art F1/IoU/AUC, including zero-shot results on OSTF, and an auxiliary stress test shows that F-DiT harmonization can degrade existing detectors.

Significance. The restoration-cost formulation is a timely and interesting alternative to discriminative forgery detectors, and the paper is generally clear. The objective in Eq. (7) is correctly specified as a weighted L2 regression, and the ablations (Tables 3 and 4) provide evidence for the pixel-space and forensic-tokenizer design choices. The inclusion of AUC as a threshold-free metric is a strength. However, the central zero-shot claim and the headline F1/IoU gains depend on a benchmark-contamination issue, an untested span assumption about the artifact-injection corruption set, and per-benchmark threshold selection. These need to be addressed before the results can be taken at face value.

major comments (3)

[Section 4.1 (Datasets)] The training corpus includes ICDAR2013 [17], while the evaluation benchmark T-IC13 [39] is a tampered version of ICDAR2013 images. Unless the original ICDAR2013 test images were explicitly excluded from training, the 'zero-shot' T-IC13 numbers (Table 1, F1=0.905) are not zero-shot: the model has seen the pristine counterparts of the test images. This is load-bearing for the abstract's zero-shot claim. Please report the overlap, re-train without ICDAR2013 (or at least without the T-IC13 source images), and update the averages and conclusions accordingly.
[Section 3.4 / Section 4.2] The detector is trained exclusively on five locally injected corruptions (Gaussian blur, JPEG, Gaussian noise, alpha-blending boundary mismatch, glyph re-rendering mismatch). The strong zero-shot claim on 'unseen text editing patterns' assumes that real tampering traces lie within the span of these five operations. No experiment tests this span: there is no evaluation on editing pipelines outside the five, no training variant with a broader corruption set, and no paired real-tampering supervision. Please add an evaluation on editing pipelines not represented in the five operations (e.g., diffusion-based inpainting, copy-move, or seamless editing), or a controlled experiment varying the corruption set, to substantiate the generalization claim.
[Section 4.1 (Implementation Details) / Table 1] F1 and IoU are computed using per-benchmark thresholds selected on each benchmark's validation set (τ=0.34 for T-IC13, τ=0.31 for DocTamper, τ=0.33 for OSTF). This makes the reported zero-shot F1/IoU dependent on benchmark-specific labels, and the headline 3.2/4.8 percentage-point gains are not threshold-free. While AUC is reported and is threshold-independent, the abstract and Section 4.2 emphasize F1/IoU. Please report threshold-robustness (e.g., precision-recall/AUC curves) and clarify how the baseline thresholds were selected, so that the comparison is consistent.

minor comments (5)

[Table 1] The DAF row appears to have a formatting issue: '0.8600.9730.759' and the DocTamper and OSTF F1/IoU values are listed as identical (0.759/0.612). Please verify the numbers and fix the spacing.
[Abstract / Section 4.2] The 3.2 and 4.8 percentage-point gains in F1 and IoU refer to the full-shot average in Table 1, not the zero-shot setting. Please state this explicitly to avoid ambiguity.
[Section 4.1] The definition of 'zero-shot' should mention that validation-selected thresholds are used per benchmark. As written, 'without benchmark-specific supervised fine-tuning' is misleading because the thresholds do use benchmark validation labels.
[Section 4.3] The stress-test analysis would be stronger with quantitative feature-space statistics (e.g., distances between cluster means or a nearest-neighbor measure) in addition to the t-SNE visualization. The 'directional movement' claim is currently qualitative.
[Throughout] T-IC13 should be defined at first use (e.g., 'Tampered-IC13, a benchmark derived from ICDAR2013'). Minor typographical errors in equations and table formatting should also be corrected.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the SC-RF objective regresses to self-injected corruption targets, and the inference score is a learned field, not a re-disclosed training label.

full rationale

The derivation chain is self-contained rather than circular. The training signal (Eq. 7) is a spatially weighted regression of v_theta to x1 - x0, where x0 is generated by Artifact Injection on authentic images and x1 is the original. The inference score (Eq. 12) is the norm of the learned instantaneous velocity on a query image. No test label, tampering annotation, or benchmark-specific supervision enters this objective; the relation between high restoration cost and real tampering is an empirical generalization claim, not an equation-level equivalence. The paper's own caveat that M_r is 'the authentic distribution induced by our training corpus and forensic priors' (Sec. 3.2) is a scope limitation, not a circular definition: it acknowledges the method measures deviation from a learned empirical distribution rather than a universal ground truth. The F1/IoU thresholds are validation-selected per benchmark, which is disclosed and standard, and the paper additionally reports threshold-free AUC that shows the same ranking (e.g., 0.933 avg AUC vs. 0.910 for the runner-up). Self-citations ([44], [45], [24]) appear only in related-work or overfitting-motivation contexts and are not load-bearing for the central derivation; there is no self-citation uniqueness theorem or ansatz smuggled via citation. The main weakness, that the five Artifact Injection operations may not span real editing traces, is an empirical coverage assumption rather than a circular reduction. Therefore score 0.

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

The paper does not introduce new physical entities or new conserved quantities. The main epistemic load is carried by the self-supervised corruption model: the detector treats 'authenticity' as the empirical distribution of the training corpus and 'deviation' as whatever Artifact Injection synthesizes. The thresholds and λ are the most consequential free parameters because they directly shape the reported F1/IoU results.

free parameters (4)

Sparsity weight λ = 20
Hyperparameter in Eq. (6) controlling the penalty on tampered regions; set by hand and directly determines the balance between authentic and anomalous pixels in the loss.
Per-benchmark thresholds τ = 0.34 (T-IC13), 0.31 (DocTamper), 0.33 (OSTF)
Validation-selected thresholds used to convert the continuous score map into binary masks; the reported F1/IoU numbers depend on these fitted values.
Frequency-gating scalar η = initialized 1.0, learned
Learnable scalar in Eq. (11) that controls how strongly spectral inconsistencies suppress attention; optimized jointly with the network.
Curriculum schedule and artifact injection hyperparameters = Stage lengths 60k/90k/150k; coverage 50%-80% / linear to 5% / 1%-5%
Chosen by hand to stabilize training; the exact artifact types, intensities, and orders are randomized but not specified quantitatively, leaving implicit free parameters.

assumptions (4)

domain assumption Training corpora (LSVT, ReCTS, ICDAR2013, ICDAR2015, ICDAR2017, TextOCR, ArT, RVL-CDIP) are authentic and tamper-free.
The target distribution M_r is defined as the empirical distribution of these images; if some contain tampering, the authenticity prior is contaminated.
domain assumption Artifact Injection perturbations approximate the residual statistics of real, unseen text editing.
The detector is trained only on synthetic blur/JPEG/noise/alpha-blending/glyph re-rendering corruptions; transfer to real tampering is assumed but not proven.
domain assumption Text region annotations are available on all training images to enable artifact injection.
The method 'randomly select[s] text regions' (Section 3.4), which requires text detection or annotations; this is not explicitly discussed as a resource requirement.
domain assumption The magnitude of the predicted velocity field at t=0 is a monotone indicator of tampering probability.
Eq. (12) converts vector magnitude into a tampering score; this assumes authentic regions have near-zero restoration cost and tampered regions have consistently larger cost.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Open-Set Visual Text Forensics via Sparse-Constraint Rectified Flow." pith.science (2026). https://pith.science/paper/MX57FBC2

@misc{pith2026260802258,
  author       = {Pith},
  title        = {Pith review of: Open-Set Visual Text Forensics via Sparse-Constraint Rectified Flow},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MX57FBC2}},
  note         = {Machine review of arXiv:2608.02258}
}

read the original abstract

Rapidly evolving Generative AI enables sophisticated visual text manipulations that increasingly evade current forensic detectors. Existing discriminative models often overfit specific forgery patterns, limiting their generalization to unseen, open-set attacks. To address this challenge, we propose a generative detector that localizes tampering by estimating the local restoration cost required to align a query image with authentic visual-text statistics, rather than by learning forgery-specific decision boundaries. Specifically, we introduce Sparse-Constraint Rectified Flow (SC-RF), a detector-oriented adaptation of Flow Matching for spatially sparse anomaly localization. We further mitigate data scarcity via self-supervised Artifact Injection and preserve high-frequency forensic traces using a pixel-space Forensic-DiT. Extensive experiments on three benchmarks show that our method achieves state-of-the-art performance, surpassing the runner-up by 3.2 and 4.8 percentage points in F1 and IoU, respectively. In particular, the proposed detector demonstrates strong zero-shot performance on challenging unseen text editing patterns. We further provide an auxiliary stress-test analysis showing that local harmonization produced by our model can weaken the statistical cues relied upon by existing detectors, offering a complementary vulnerability-analysis perspective.

Figures

Figures reproduced from arXiv: 2608.02258 by the authors.

**Figure 2.** Overview of our proposed framework. (a) Data Preparation: We synthesize the source distribution [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Qualitative comparison of tampering localization results on challenging samples from OSTF, DocTamper, and T-IC13 [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: t-SNE visualization of feature distributions. (a) In [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

49 extracted references · 2 linked inside Pith

[17]

Dimosthenis Karatzas, Faisal Shafait, Seiichi Uchida, Masakazu Iwamura, Lluis Gomez i Bigorda, Sergi Robles Mestre, Joan Mas, David Fernández Mota, Jon Almazán, and Lluís-Pere de las Heras. 2013. ICDAR 2013 Robust Reading Competition. InProceedings of the International Conference on Document Analysis and Recognition (ICDAR). 1484–1493

2013
[39]

Yuxin Wang, Hongtao Xie, Mengting Xing, Jing Wang, Shenggao Zhu, and Yong- dong Zhang. 2022. Detecting Tampered Scene Text in the Wild. InComputer Vision – ECCV 2022: 17th European Conference, Tel A viv, Israel, October 23–27, 2022, Proceedings, Part XXVIII (Lecture Notes in Computer Science, Vol. 13688), Shai Avidan, Gabriel J. Brostow, Moustapha Cissé, ...

2022
[1]

Albergo and Eric Vanden-Eijnden

Michael S. Albergo and Eric Vanden-Eijnden. 2023. Building Normalizing Flows with Stochastic Interpolants. InInternational Conference on Learning Representa- tions (ICLR)

2023
[2]

Jinwon An and Sungzoon Cho. 2015. Variational Autoencoder Based Anomaly Detection Using Reconstruction Probability.Special Lecture on IE2, 1 (2015), 1–18

2015
[3]

Belhassen Bayar and Matthew C. Stamm. 2016. A Deep Learning Approach to Universal Image Manipulation Detection Using a New Convolutional Layer. InProceedings of the 4th ACM Workshop on Information Hiding and Multimedia Security (IH&MMSec). ACM, 5–10

2016
[4]

Abhijit Bendale and Terrance E. Boult. 2016. Towards Open Set Deep Networks. InProceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 1563–1572

2016
[5]

Xiuli Bi, Yang Wei, Bin Xiao, and Weisheng Li. 2019. RRU-Net: The Ringed Resid- ual U-Net for Image Splicing Forgery Detection. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops

2019
[6]

Liang-Chieh Chen, Yukun Zhu, George Papandreou, Florian Schroff, and Hartwig Adam. 2018. Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation. InProceedings of the European Conference on Computer Vision (ECCV). 833–851

2018

Show all 49 references

[7]

Xinru Chen, Chengbo Dong, Jiaqi Ji, Juan Cao, and Xirong Li. 2021. Image Manipulation Detection by Multi-View Multi-Scale Supervision. InProceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). 14165–14173

2021
[8]

Chee Kheng Chng, Errui Ding, Jingtuo Liu, Dimosthenis Karatzas, Chee Seng Chan, Lianwen Jin, Yuliang Liu, Yipeng Sun, Chun Chet Ng, Canjie Luo, Zihan Ni, ChuanMing Fang, Shuaitao Zhang, and Junyu Han. 2019. ICDAR2019 Robust Reading Challenge on Arbitrary-Shaped Text – RRC-ArT....

2019
[9]

Francisco Cruz, Nicolas Sidere, Mickaël Coustaty, Vincent Poulain D’Andecy, and Jean-Marc Ogier. 2017. Local Binary Patterns for Document Forgery Detection. InProceedings of the 14th IAPR International Conference on Document Analysis and Recognition (ICDAR). 1223–1228

2017
[10]

Prafulla Dhariwal and Alexander Quinn Nichol. 2021. Diffusion Models Beat GANs on Image Synthesis. InAdvances in Neural Information Processing Systems (NeurIPS). 8780–8794

2021
[11]

Junxian Duan, Hao Sun, Fan Ji, Kai Zhou, Zhiyong Wang, Huaibo Huang, and Lianwen Jin. 2025. RealDTT: Towards a Comprehensive Real-World Dataset for Tampered Text Detection.International Journal of Computer Vision133, 10 (2025), 6993–7011

2025
[12]

Joel Frank, Thorsten Eisenhofer, Lea Schönherr, Asja Fischer, Dorothea Kolossa, and Thorsten Holz. 2020. Leveraging Frequency Analysis for Deep Fake Image Recognition. InProceedings of the International Conference on Machine Learning (ICML). 3247–3258

2020
[13]

Fridrich and Jan Kodovský

Jessica J. Fridrich and Jan Kodovský. 2012. Rich Models for Steganalysis of Digital Images.IEEE Transactions on Information Forensics and Security7, 3 (2012), 868–882

2012
[14]

Dong Gong, Lingqiao Liu, Vuong Le, Budhaditya Saha, Moussa Mansour, Svetha Venkatesh, and Anton van den Hengel. 2019. Memorizing Normality to Detect Anomaly: Memory-Augmented Deep Autoencoder for Unsupervised Anomaly Detection. InProceedings of the IEEE/CVF International Confe...

2019
[15]

Harley, Alex Ufkes, and Konstantinos G

Adam W. Harley, Alex Ufkes, and Konstantinos G. Derpanis. 2015. Evaluation of Deep Convolutional Nets for Document Image Classification and Retrieval. In Proceedings of the International Conference on Document Analysis and Recognition (ICDAR). 991–995

2015
[16]

Ghosh, Andrew D

Dimosthenis Karatzas, Lluis Gomez-Bigorda, Anguelos Nicolaou, Suman K. Ghosh, Andrew D. Bagdanov, Masakazu Iwamura, Jiri Matas, Lukás Neumann, Vijay Ramaseshan Chandrasekhar, Shijian Lu, Faisal Shafait, Seiichi Uchida, and Ernest Valveny. 2015. ICDAR 2015 Competition on Robust...

2015
[18]

Kingma and Prafulla Dhariwal

Durk P. Kingma and Prafulla Dhariwal. 2018. Glow: Generative Flow with Invertible 1x1 Convolutions. InAdvances in Neural Information Processing Systems (NeurIPS)

2018
[19]

Xin Liao, Siliang Chen, Jiaxin Chen, Tianyi Wang, and Xiehua Li. 2023. CTP-Net: Character Texture Perception Network for Document Image Forgery Localization. arXiv preprint arXiv:2308.02158(2023)

2023 arXiv
[20]

Yaron Lipman, Ricky T. Q. Chen, Heli Ben-Hamu, Maximilian Nickel, and Matthew Le. 2023. Flow Matching for Generative Modeling. InInternational Conference on Learning Representations (ICLR)

2023
[21]

Qiang Liu. 2022. Rectified Flow: A Marginal Preserving Approach to Optimal Transport.arXiv preprint arXiv:2209.14577(2022)

2022 arXiv
[22]

Xingchao Liu, Chengyue Gong, and Qiang Liu. 2023. Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow. InThe Eleventh International Conference on Learning Representations (ICLR 2023), Kigali, Rwanda, May 1–5, 2023. OpenReview.net. https://openrev...

2023
[23]

Xiaohong Liu, Yaojie Liu, Jun Chen, and Xiaoming Liu. 2022. PSCC-Net: Progres- sive Spatio-Channel Correlation Network for Image Manipulation Detection and Localization.IEEE Transactions on Circuits and Systems for Video Technology32, 11 (2022), 7505–7517

2022
[24]

Dongliang Luo, Yuliang Liu, Rui Yang, Xianjin Liu, Jishen Zeng, Yu Zhou, and Xiang Bai. 2025. Toward Real Text Manipulation Detection: New Dataset and New Solution.Pattern Recognition157 (2025), 110828

2025
[25]

Nibal Nayef, Fei Yin, Imen Bizid, Hyunsoo Choi, Yuan Feng, Dimosthenis Karatzas, Zhenbo Luo, Umapada Pal, Christophe Rigaud, Joseph Chazalon, Wafa Khlif, Muhammad Muzzamil Luqman, Jean-Christophe Burie, Cheng-Lin Liu, and Jean- Marc Ogier. 2017. ICDAR2017 Robust Reading Challe...

2017
[26]

Alexander Quinn Nichol and Prafulla Dhariwal. 2021. Improved Denoising Diffusion Probabilistic Models. InProceedings of the International Conference on Machine Learning (ICML). 8162–8171

2021
[27]

Chenfan Qu, Chongyu Liu, Yuliang Liu, Xinhong Chen, Dezhi Peng, Fengjun Guo, and Lianwen Jin. 2023. Towards Robust Tampered Text Detection in Document Image: New Dataset and New Solution. InIEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2023), Vancouver, ...

2023
[28]

Chenfan Qu, Yiwu Zhong, Fengjun Guo, and Lianwen Jin. 2025. Revisiting Tampered Scene Text Detection in the Era of Generative AI. InProceedings of the AAAI Conference on Artificial Intelligence, Vol. 39. 694–702

2025
[29]

Danilo Rezende and Shakir Mohamed. 2015. Variational Inference with Normal- izing Flows. InProceedings of the International Conference on Machine Learning (ICML). 1530–1538

2015
[30]

Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Björn Ommer. 2022. High-Resolution Image Synthesis with Latent Diffusion Models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 10674–10685

2022
[31]

Olaf Ronneberger, Philipp Fischer, and Thomas Brox. 2015. U-Net: Convolutional Networks for Biomedical Image Segmentation. InProceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention (MIC- CAI). 234–241

2015
[32]

Mayu Sakurada and Takehisa Yairi. 2014. Anomaly Detection Using Autoencoders with Nonlinear Dimensionality Reduction. InProceedings of the MLSDA Workshop. 4–11

2014
[33]

Waldstein, and Georg Langs

Thomas Schlegl, Philipp Seeböck, Sebastian M. Waldstein, and Georg Langs. 2019. Fast Unsupervised Anomaly Detection with Generative Adversarial Networks. Medical Image Analysis54 (2019), 30–44

2019
[34]

Amanpreet Singh, Guan Pang, Mandy Toh, Jing Huang, Wojciech Galuba, and Tal Hassner. 2021. TextOCR: Towards Large-Scale End-to-End Reasoning for Arbitrary-Shaped Scene Text. InProceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition (CVPR). 8802–8812

2021
[35]

Yipeng Sun, Dimosthenis Karatzas, Chee Seng Chan, Lianwen Jin, Zihan Ni, Chee Kheng Chng, Yuliang Liu, Canjie Luo, Chun Chet Ng, Junyu Han, Errui Ding, and Jingtuo Liu. 2019. ICDAR 2019 Competition on Large-Scale Street View Text with Partial Labeling – RRC-LSVT. InProceedings...

2019
[36]

Yuchuan Tian, Zhijun Tu, Hanting Chen, Jie Hu, Chao Xu, and Yunhe Wang. 2024. U-DiTs: Downsample Tokens in U-Shaped Diffusion Transformers. InAdvances in Neural Information Processing Systems (NeurIPS)

2024
[37]

Gomez, Łukasz Kaiser, and Illia Polosukhin

Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention Is All You Need. InAdvances in Neural Information Processing Systems (NeurIPS)

2017
[38]

Sheng-Yu Wang, Oliver Wang, Richard Zhang, Andrew Owens, and Alexei A. Efros. 2020. CNN-generated images are surprisingly easy to spot... for now. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 8695–8704

2020
[40]

Yue Wu, Wael AbdAlmageed, and Premkumar Natarajan. 2019. ManTra-Net: Manipulation Tracing Network for Detection and Localization of Image Forgeries with Anomalous Features. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 9543–9552. M...

2019
[41]

Schmon, Chris G

James Wyatt, Alex Leach, Samuel M. Schmon, Chris G. Willcocks, and An- drew Gordon Wilson. 2022. AnoDDPM: Anomaly Detection with Denoising Dif- fusion Probabilistic Models Using Simplex Noise. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition ...

2022
[42]

Alvarez, and Ping Luo

Enze Xie, Wenhai Wang, Zhiding Yu, Anima Anandkumar, Jose M. Alvarez, and Ping Luo. 2021. SegFormer: Simple and Efficient Design for Semantic Segmenta- tion with Transformers. InAdvances in Neural Information Processing Systems, Vol. 34. 12077–12090

2021
[43]

Wenbo Xu, Junwei Luo, Chuntao Zhu, Wei Lu, Jinhua Zeng, Shaopei Shi, and Cong Lin. 2022. Document Images Forgery Localization Using a Two-Stream Network.International Journal of Intelligent Systems37, 8 (2022), 5272–5289

2022
[44]

Jiangling Zhang, Shuxuan Gao, Bofan Liu, Siqiang Feng, Jirui Huang, Yaxiong Chen, and Ziyu Chen. 2026. Detecting AI-Generated Forgeries via Iterative Manifold Deviation Amplification. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). I...

2026
[45]

Jiangling Zhang, Weijie Zhu, Jirui Huang, and Yaxiong Chen. 2025. LAMM-ViT: AI Face Detection via Layer-Aware Modulation of Region-Guided Attention. InECAI 2025 – 28th European Conference on Artificial Intelligence, Bologna, Italy, October 25–30, 2025 (Frontiers in Artificial ...

2025
[46]

Rui Zhang, Mingkun Yang, Xiang Bai, Baoguang Shi, Dimosthenis Karatzas, Shijian Lu, C. V. Jawahar, Yongsheng Zhou, Qianyi Jiang, Qi Song, Nan Li, Kai Zhou, Lei Wang, Dong Wang, and Minghui Liao. 2019. ICDAR 2019 Robust Reading Challenge on Reading Chinese Text on Signboard. In...

2019
[47]

Xinyu Zhang, Ning Li, Juncheng Li, Zhen Wang, and Rui Huang. 2023. Un- supervised Surface Anomaly Detection with Diffusion Probabilistic Model. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). 6782–6791

2023
[48]

Morariu, and Larry S

Peng Zhou, Xintong Han, Vlad I. Morariu, and Larry S. Davis. 2018. Learning Rich Features for Image Manipulation Detection. InProceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 1053–1061

2018
[49]

Peng Zhuang, Hong Li, Shaohua Tan, Bin Li, and Jianping Huang. 2021. Image Tampering Localization Using a Dense Fully Convolutional Network.IEEE Transactions on Information Forensics and Security16 (2021), 2986–2999

2021

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.

[1] [17]

Dimosthenis Karatzas, Faisal Shafait, Seiichi Uchida, Masakazu Iwamura, Lluis Gomez i Bigorda, Sergi Robles Mestre, Joan Mas, David Fernández Mota, Jon Almazán, and Lluís-Pere de las Heras. 2013. ICDAR 2013 Robust Reading Competition. InProceedings of the International Conference on Document Analysis and Recognition (ICDAR). 1484–1493

2013

[2] [39]

Yuxin Wang, Hongtao Xie, Mengting Xing, Jing Wang, Shenggao Zhu, and Yong- dong Zhang. 2022. Detecting Tampered Scene Text in the Wild. InComputer Vision – ECCV 2022: 17th European Conference, Tel A viv, Israel, October 23–27, 2022, Proceedings, Part XXVIII (Lecture Notes in Computer Science, Vol. 13688), Shai Avidan, Gabriel J. Brostow, Moustapha Cissé, ...

2022

[3] [1]

Albergo and Eric Vanden-Eijnden

Michael S. Albergo and Eric Vanden-Eijnden. 2023. Building Normalizing Flows with Stochastic Interpolants. InInternational Conference on Learning Representa- tions (ICLR)

2023

[4] [2]

Jinwon An and Sungzoon Cho. 2015. Variational Autoencoder Based Anomaly Detection Using Reconstruction Probability.Special Lecture on IE2, 1 (2015), 1–18

2015

[5] [3]

Belhassen Bayar and Matthew C. Stamm. 2016. A Deep Learning Approach to Universal Image Manipulation Detection Using a New Convolutional Layer. InProceedings of the 4th ACM Workshop on Information Hiding and Multimedia Security (IH&MMSec). ACM, 5–10

2016

[6] [4]

Abhijit Bendale and Terrance E. Boult. 2016. Towards Open Set Deep Networks. InProceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 1563–1572

2016

[7] [5]

Xiuli Bi, Yang Wei, Bin Xiao, and Weisheng Li. 2019. RRU-Net: The Ringed Resid- ual U-Net for Image Splicing Forgery Detection. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops

2019

[8] [6]

Liang-Chieh Chen, Yukun Zhu, George Papandreou, Florian Schroff, and Hartwig Adam. 2018. Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation. InProceedings of the European Conference on Computer Vision (ECCV). 833–851

2018