pith. sign in

arxiv: 2509.05753 · v2 · submitted 2025-09-06 · 💻 cs.CR · cs.AI· cs.CV

Tell-Tale Watermarks for Explanatory Reasoning in Synthetic Media Forensics

Ching-Chun Chang , Isao Echizen This is my paper

Pith reviewed 2026-05-18 17:45 UTC · model grok-4.3

classification 💻 cs.CR cs.AIcs.CV
keywords tell-tale watermarkingsynthetic media forensicsdigital image transformationsexplanatory reasoningtraceabilityfidelitysynchronicity
0
0 comments X

The pith

Tell-tale watermarks embedded in synthetic media evolve with edits to reveal the sequence of transformations applied.

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

The paper develops a watermarking method for digital images that responds to different kinds of changes in an interpretable way rather than staying fixed or disappearing. These markers are tuned separately for content alterations, color shifts, and viewpoint changes so that the pattern of their degradation points back to the specific edits performed. By combining the observed responses, the method reconstructs the most likely chain of operations that produced the final image. This matters for forensics because it can help distinguish innocent adjustments from deliberate attempts to mislead viewers about the origin or content of the media.

Core claim

Tell-tale watermarks are tailored to distinct transformation classes so that they leave interpretable traces when the carrier media undergoes semantic, photometric, or geometric changes. These watermarks function as reference clues that evolve under the same dynamics as the media itself, allowing explanatory reasoning to infer the most plausible account of composite transformations across the lifecycle of synthetic media.

What carries the argument

Tell-tale watermarks: reference signals embedded in the media that respond to transformations in a manner that is neither fully robust nor fully fragile but instead produces class-specific, readable traces.

Load-bearing premise

Watermarks can be designed so their changes under semantic, photometric, and geometric edits stay distinct enough to support reliable inference about which combination of edits occurred.

What would settle it

An experiment in which two different sequences of transformations produce identical or statistically indistinguishable patterns in the extracted watermarks, so that no unique explanatory account can be selected.

Figures

Figures reproduced from arXiv: 2509.05753 by Ching-Chun Chang, Isao Echizen.

Figure 1
Figure 1. Figure 1: Overview of tell-tale watermarking and explanatory reasoning in response to semantic, photometric and geometric transformations. [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Architecture of the encoder and decoder neural networks, with implementation details for the U-Net module and the residual attention module. [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Demonstration of tell-tale watermarks under different transformation chains. [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Fidelity evaluation of watermark insertion based on distortion metrics [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗
Figure 6
Figure 6. Figure 6: Traceability evaluation of semantic reasoning based on IoU between [PITH_FULL_IMAGE:figures/full_fig_p010_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Traceability evaluation of photometric and geometric reasoning based on deviations between estimated and true parameter values. [PITH_FULL_IMAGE:figures/full_fig_p011_7.png] view at source ↗
read the original abstract

The rise of synthetic media has blurred the boundary between reality and fabrication under the evolving power of artificial intelligence, fueling an infodemic that erodes public trust in cyberspace. For digital imagery, a multitude of editing applications further complicates the forensic analysis, including semantic edits that alter content, photometric adjustments that recalibrate colour characteristics, and geometric projections that reshape viewpoints. Collectively, these transformations manipulate and control perceptual interpretation of digital imagery. This susceptibility calls for forensic enquiry into reconstructing the chain of events, thereby revealing deeper evidential insight into the presence or absence of criminal intent. This study seeks to address an inverse problem of tracing the underlying generation chain that gives rise to the observed synthetic media. A tell-tale watermarking system is developed for explanatory reasoning over the nature and extent of transformations across the lifecycle of synthetic media. Tell-tale watermarks are tailored to different classes of transformations, responding in a manner that is neither strictly robust nor fragile but instead interpretable. These watermarks function as reference clues that evolve under the same transformation dynamics as the carrier media, leaving interpretable traces when subjected to transformations. Explanatory reasoning is then performed to infer the most plausible account across the combinatorial parameter space of composite transformations. Experimental evaluations demonstrate the validity of tell-tale watermarking with respect to fidelity, synchronicity and traceability.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The manuscript presents a tell-tale watermarking system for explanatory reasoning in synthetic media forensics. It develops watermarks tailored to different classes of transformations (semantic, photometric, geometric) that respond in an interpretable manner, evolving with the carrier media to leave traces. Explanatory reasoning is performed to infer the most plausible account of the transformation sequence across the combinatorial parameter space. The paper reports experimental evaluations demonstrating validity with respect to fidelity, synchronicity, and traceability.

Significance. Should the proposed tell-tale watermarking approach prove effective in practice, it would offer a valuable contribution to the field by addressing the inverse problem of tracing the generation chain of synthetic media. This could enhance forensic capabilities in distinguishing between different types of edits and reconstructing their order, which is crucial for understanding the extent and nature of manipulations in digital imagery. The interpretable response design is a promising departure from conventional watermarking strategies.

major comments (2)
  1. The abstract asserts experimental validation on fidelity, synchronicity, and traceability but supplies no quantitative results, baselines, error analysis, or methodology details, leaving the support for the claim unassessable. This is load-bearing for evaluating the soundness of the central claims.
  2. The explanatory reasoning component claims that class-specific watermark responses support reliable inference to recover the most plausible transformation sequence from the combinatorial space, but no concrete inference procedure (e.g., optimization or probabilistic model) or validation against interaction effects under composition (such as non-commutative semantic-geometric interactions) is described. This directly affects the traceability claim.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their positive evaluation of the significance of our work and for the constructive major comments. We address each point below and indicate the revisions planned for the next version of the manuscript.

read point-by-point responses

  1. Referee: The abstract asserts experimental validation on fidelity, synchronicity, and traceability but supplies no quantitative results, baselines, error analysis, or methodology details, leaving the support for the claim unassessable. This is load-bearing for evaluating the soundness of the central claims.

    Authors: We agree that the abstract, as a concise summary, does not include specific quantitative results, baselines, or methodological details, which limits immediate assessment of the claims. The full manuscript contains these elements in the dedicated experimental evaluation section. To address the referee's concern, we will revise the abstract to incorporate concise quantitative highlights and a brief reference to the evaluation methodology. revision: yes

  2. Referee: The explanatory reasoning component claims that class-specific watermark responses support reliable inference to recover the most plausible transformation sequence from the combinatorial space, but no concrete inference procedure (e.g., optimization or probabilistic model) or validation against interaction effects under composition (such as non-commutative semantic-geometric interactions) is described. This directly affects the traceability claim.

    Authors: The manuscript describes explanatory reasoning as using the interpretable, class-specific watermark responses to identify the most plausible transformation sequence over the combinatorial space. We acknowledge that an explicit description of the inference procedure and targeted validation for interaction effects under composition would strengthen the traceability claims. We will add a new subsection detailing the inference procedure and include additional experiments validating robustness to non-commutative interactions between transformation classes. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The manuscript proposes a tell-tale watermarking design for tracing composite transformations in synthetic media, with watermarks tailored to semantic, photometric, and geometric classes to support explanatory inference. No equations, derivations, or fitted parameters are described that reduce the claimed fidelity, synchronicity, or traceability to self-referential definitions or prior fitted inputs. The central claims rest on new design choices and experimental validation rather than any self-citation chain or ansatz smuggled from prior work. The inference procedure over combinatorial spaces is presented as part of the novel contribution, not presupposed by construction. This is a self-contained proposal against external benchmarks with no load-bearing circular steps.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The approach depends on domain assumptions about transformation categories and likely free parameters for watermark response tuning, with no invented entities or machine-checked elements visible.

free parameters (1)
  • watermark response parameters
    Parameters controlling how watermarks react to each transformation class are required to achieve the desired interpretable traces and are not derived from first principles.
axioms (1)
  • domain assumption Transformations fall into distinct semantic, photometric, and geometric classes that can be addressed separately by tailored watermarks.
    This classification underpins the tailoring strategy and is invoked to enable the explanatory reasoning step.

pith-pipeline@v0.9.0 · 5767 in / 1277 out tokens · 46351 ms · 2026-05-18T17:45:51.088758+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

73 extracted references · 73 canonical work pages

  1. [1]

    When seeing isn’t believing [multimedia authentication technologies],

    B. Zhu, M. Swanson, and A. Tewfik, “When seeing isn’t believing [multimedia authentication technologies],”IEEE Signal Process. Mag., vol. 21, no. 2, pp. 40–49, 2004

    work page 2004

  2. [2]

    Can people identify original and manipulated photos of real-world scenes?

    S. J. Nightingale, K. A. Wade, and D. G. Watson, “Can people identify original and manipulated photos of real-world scenes?”Cogn. Res. Princ. Implic., vol. 2, no. 1, pp. 1–21, 2017

    work page 2017

  3. [3]

    Deepfakes and the new disinformation war,

    R. Chesney and D. Citron, “Deepfakes and the new disinformation war,” F oreign Aff., vol. 98, no. 1, pp. 147–155, 2019

    work page 2019

  4. [4]

    Will deepfakes do deep damage?

    S. Greengard, “Will deepfakes do deep damage?”Commun. ACM, vol. 63, no. 1, pp. 17–19, 2019

    work page 2019

  5. [5]

    Generative adversarial networks,

    I. Goodfellowet al., “Generative adversarial networks,”Commun. ACM, vol. 63, no. 11, pp. 139–144, 2020

    work page 2020

  6. [6]

    The creation and detection of deepfakes: A survey,

    Y . Mirsky and W. Lee, “The creation and detection of deepfakes: A survey,”ACM Comput. Surv., vol. 54, no. 1, pp. 1–41, 2021

    work page 2021

  7. [7]

    Real-time expression transfer for facial reenactment,

    J. Thies, M. Zollhöfer, M. Nießner, L. Valgaerts, M. Stamminger, and C. Theobalt, “Real-time expression transfer for facial reenactment,” ACM Trans. Graph., vol. 34, no. 6, pp. 1–14, 2015

    work page 2015

  8. [8]

    Neural photo editing with introspective adversarial networks,

    A. Brock, T. Lim, J. M. Ritchie, and N. Weston, “Neural photo editing with introspective adversarial networks,” inProc. Int. Conf. Learn. Represent. (ICLR), Toulon, France, 2017, pp. 1–15

    work page 2017

  9. [9]

    Deep video portraits,

    H. Kimet al., “Deep video portraits,”ACM Trans. Graph., vol. 37, no. 4, pp. 1–14, 2018

    work page 2018

  10. [10]

    Face2Face: Real-time face capture and reenactment of RGB videos,

    J. Thies, M. Zollhöfer, M. Stamminger, C. Theobalt, and M. Nießner, “Face2Face: Real-time face capture and reenactment of RGB videos,” Commun. ACM, vol. 62, no. 1, pp. 96–104, 2018

    work page 2018

  11. [11]

    X2Face: A network for controlling face generation using images, audio, and pose codes,

    O. Wiles, A. S. Koepke, and A. Zisserman, “X2Face: A network for controlling face generation using images, audio, and pose codes,” in Proc. Eur . Conf. Comput. Vis. (ECCV), Munich, Germany, 2018, pp. 690–706

    work page 2018

  12. [12]

    Neural rendering and reenactment of human actor videos,

    L. Liuet al., “Neural rendering and reenactment of human actor videos,” ACM Trans. Graph., vol. 38, no. 5, pp. 1–14, 2019, Art. no. 139

    work page 2019

  13. [13]

    Neural voice puppetry: Audio-driven facial reenactment,

    J. Thies, M. Elgharib, A. Tewari, C. Theobalt, and M. Nießner, “Neural voice puppetry: Audio-driven facial reenactment,” inProc. Eur . Conf. Comput. Vis. (ECCV), Glasgow, UK, 2020, pp. 716–731

    work page 2020

  14. [14]

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

    C. Sahariaet al., “Photorealistic text-to-image diffusion models with deep language understanding,” inProc. Int. Conf. Neural Inf. Process. Syst. (NeurIPS), vol. 35, New Orleans, LA, USA, 2022, pp. 36 479– 36 494

    work page 2022

  15. [15]

    The spreading of misinformation online,

    M. Del Vicarioet al., “The spreading of misinformation online,”Proc. Natl. Acad. Sci. USA, vol. 113, no. 3, pp. 554–559, 2016

    work page 2016

  16. [16]

    Beyond misinformation: Understanding and coping with the “post-truth

    S. Lewandowsky, U. K. H. Ecker, and J. Cook, “Beyond misinformation: Understanding and coping with the “post-truth” era,”J. Appl. Res. Mem. Cogn., vol. 6, no. 4, pp. 353–369, 2017

    work page 2017

  17. [17]

    The science of fake news,

    D. M. J. Lazeret al., “The science of fake news,”Science, vol. 359, no. 6380, pp. 1094–1096, 2018

    work page 2018

  18. [18]

    The spread of true and false news online,

    S. V osoughi, D. Roy, and S. Aral, “The spread of true and false news online,”Science, vol. 359, no. 6380, pp. 1146–1151, 2018

    work page 2018

  19. [19]

    Next-generation digital forensics,

    G. G. Richard and V . Roussev, “Next-generation digital forensics,” Commun. ACM, vol. 49, no. 2, pp. 76–80, 2006

    work page 2006

  20. [20]

    Digital image forensics,

    J. Fridrich, “Digital image forensics,”IEEE Signal Process. Mag., vol. 26, no. 2, pp. 26–37, 2009

    work page 2009

  21. [21]

    Image forgery detection,

    H. Farid, “Image forgery detection,”IEEE Signal Process. Mag., vol. 26, no. 2, pp. 16–25, 2009

    work page 2009

  22. [22]

    Vision of the unseen: Current trends and challenges in digital image and video forensics,

    A. Rocha, W. Scheirer, T. Boult, and S. Goldenstein, “Vision of the unseen: Current trends and challenges in digital image and video forensics,”ACM Comput. Surv., vol. 43, no. 4, pp. 1–42, 2011

    work page 2011

  23. [23]

    Forensic detection of image manipulation using statistical intrinsic fingerprints,

    M. C. Stamm and K. R. Liu, “Forensic detection of image manipulation using statistical intrinsic fingerprints,”IEEE Trans. Inf. F orensics Secur ., vol. 5, no. 3, pp. 492–506, 2010

    work page 2010

  24. [24]

    Learning rich features for image manipulation detection,

    P. Zhou, X. Han, V . I. Morariu, and L. S. Davis, “Learning rich features for image manipulation detection,” inProc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), Salt Lake City, UT, USA, 2018, pp. 1053–1061

    work page 2018

  25. [25]

    Defending against neural fake news,

    R. Zellerset al., “Defending against neural fake news,” inProc. Int. Conf. Neural Inf. Process. Syst. (NeurIPS), vol. 32, Vancouver, BC, Canada, 2019, pp. 9054–9065

    work page 2019

  26. [26]

    Do GANs leave artificial fingerprints?

    F. Marra, D. Gragnaniello, L. Verdoliva, and G. Poggi, “Do GANs leave artificial fingerprints?” inProc. IEEE Conf. Multimedia Inf. Process. Retr . (MIPR), San Jose, CA, USA, 2019, pp. 506–511

    work page 2019

  27. [27]

    Computational sensor fingerprints,

    P. Korus and N. Memon, “Computational sensor fingerprints,”IEEE Trans. Inf. F orensics Secur ., vol. 17, pp. 2508–2523, 2022

    work page 2022

  28. [28]

    De- tectGPT: Zero-shot machine-generated text detection using probability curvature,

    E. Mitchell, Y . Lee, A. Khazatsky, C. D. Manning, and C. Finn, “De- tectGPT: Zero-shot machine-generated text detection using probability curvature,” inProc. Int. Conf. Mach. Learn. (ICML), vol. 202, Honolulu, HI, USA, 2023, pp. 24 950–24 962. 13

    work page 2023

  29. [29]

    In ictu oculi: Exposing AI created fake videos by detecting eye blinking,

    Y . Li, M.-C. Chang, and S. Lyu, “In ictu oculi: Exposing AI created fake videos by detecting eye blinking,” inProc. IEEE Int. Workshop Inf. F orensics Secur . (WIFS), Hong Kong, China, 2018, pp. 1–7

    work page 2018

  30. [30]

    Exposing deep fakes using inconsistent head poses,

    X. Yang, Y . Li, and S. Lyu, “Exposing deep fakes using inconsistent head poses,” inProc. IEEE Int. Conf. Acoust. Speech Signal Process. (ICASSP), Brighton, UK, 2019, pp. 8261–8265

    work page 2019

  31. [31]

    FaceForensics++: Learning to detect manipulated facial images,

    A. Rössler, D. Cozzolino, L. Verdoliva, C. Riess, J. Thies, and M. Niess- ner, “FaceForensics++: Learning to detect manipulated facial images,” inProc. IEEE/CVF Int. Conf. Comput. Vis. (ICCV), Seoul, Korea, 2019, pp. 1–11

    work page 2019

  32. [32]

    FakeCatcher: Detection of synthetic portrait videos using biological signals,

    U. A. Ciftci, I. Demir, and L. Yin, “FakeCatcher: Detection of synthetic portrait videos using biological signals,”IEEE Trans. Pattern Anal. Mach. Intell., pp. 1–1, 2020

    work page 2020

  33. [33]

    Face X-ray for more general face forgery detection,

    L. Liet al., “Face X-ray for more general face forgery detection,” in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), Seattle, W A, USA, 2020, pp. 5000–5009

    work page 2020

  34. [34]

    Face forensics in the wild,

    T. Zhou, W. Wang, Z. Liang, and J. Shen, “Face forensics in the wild,” inProc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), Nashville, TN, USA, 2021, pp. 5774–5784

    work page 2021

  35. [35]

    Lips don’t lie: A generalisable and robust approach to face forgery detection,

    A. Haliassos, K. V ougioukas, S. Petridis, and M. Pantic, “Lips don’t lie: A generalisable and robust approach to face forgery detection,” inProc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), Nashville, TN, USA, 2021, pp. 5037–5047

    work page 2021

  36. [36]

    Detecting deepfakes with self-blended images,

    K. Shiohara and T. Yamasaki, “Detecting deepfakes with self-blended images,” inProc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), New Orleans, LA, USA, 2022, pp. 18 699–18 708

    work page 2022

  37. [37]

    TruFor: Leveraging all-round clues for trustworthy image forgery detection and localization,

    F. Guillaro, D. Cozzolino, A. Sud, N. Dufour, and L. Verdoliva, “TruFor: Leveraging all-round clues for trustworthy image forgery detection and localization,” inProc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), Vancouver, BC, Canada, 2023, pp. 20 606–20 615

    work page 2023

  38. [38]

    The trustworthy digital camera: Restoring credibility to the photographic image,

    G. Friedman, “The trustworthy digital camera: Restoring credibility to the photographic image,”IEEE Trans. Consum. Electron., vol. 39, no. 4, pp. 905–910, 1993

    work page 1993

  39. [39]

    Multimedia data- embedding and watermarking technologies,

    M. D. Swanson, M. Kobayashi, and A. H. Tewfik, “Multimedia data- embedding and watermarking technologies,”Proc. IEEE, vol. 86, no. 6, pp. 1064–1087, 1998

    work page 1998

  40. [40]

    Information hiding—A survey,

    F. Petitcolas, R. Anderson, and M. Kuhn, “Information hiding—A survey,”Proc. IEEE, vol. 87, no. 7, pp. 1062–1078, 1999

    work page 1999

  41. [41]

    Multimedia watermarking techniques,

    F. Hartung and M. Kutter, “Multimedia watermarking techniques,”Proc. IEEE, vol. 87, no. 7, pp. 1079–1107, 1999

    work page 1999

  42. [42]

    The use of watermarks in the protection of digital multimedia products,

    G. V oyatzis and I. Pitas, “The use of watermarks in the protection of digital multimedia products,”Proc. IEEE, vol. 87, no. 7, pp. 1197–1207, 1999

    work page 1999

  43. [43]

    Watermarking as communications with side information,

    I. J. Cox, M. L. Miller, and A. L. McKellips, “Watermarking as communications with side information,”Proc. IEEE, vol. 87, no. 7, pp. 1127–1141, 1999

    work page 1999

  44. [44]

    Cyber vaccine for deepfake immunity,

    C.-C. Chang, H. H. Nguyen, J. Yamagishi, and I. Echizen, “Cyber vaccine for deepfake immunity,”IEEE Access, vol. 11, pp. 105 027– 105 039, 2023

    work page 2023

  45. [45]

    Secure spread spectrum watermarking for multimedia,

    I. J. Cox, J. Kilian, F. T. Leighton, and T. Shamoon, “Secure spread spectrum watermarking for multimedia,”IEEE Trans. Image Process., vol. 6, no. 12, pp. 1673–1687, 1997

    work page 1997

  46. [46]

    A DCT-domain system for robust image watermarking,

    M. Barni, F. Bartolini, V . Cappellini, and A. Piva, “A DCT-domain system for robust image watermarking,”Signal Process., vol. 66, no. 3, pp. 357–372, 1998

    work page 1998

  47. [47]

    An SVD-based watermarking scheme for protecting rightful ownership,

    R. Liu and T. Tan, “An SVD-based watermarking scheme for protecting rightful ownership,”IEEE Trans. Multimed., vol. 4, no. 1, pp. 121–128, 2002

    work page 2002

  48. [48]

    Distortion agnostic deep watermarking,

    X. Luo, R. Zhan, H. Chang, F. Yang, and P. Milanfar, “Distortion agnostic deep watermarking,” inProc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), Seattle, W A, USA, 2020, pp. 13 545–13 554

    work page 2020

  49. [49]

    A watermark for large language models,

    J. Kirchenbauer, J. Geiping, Y . Wen, J. Katz, I. Miers, and T. Goldstein, “A watermark for large language models,” inProc. Int. Conf. Mach. Learn. (ICML), vol. 202, Honolulu, HI, USA, 2023, pp. 17 061–17 084

    work page 2023

  50. [50]

    The stable signature: Rooting watermarks in latent diffusion models,

    P. Fernandez, G. Couairon, H. Jégou, M. Douze, and T. Furon, “The stable signature: Rooting watermarks in latent diffusion models,” in Proc. IEEE/CVF Int. Conf. Comput. Vis. (ICCV), Paris, France, 2023, pp. 22 409–22 420

    work page 2023

  51. [51]

    Proactive detection of voice cloning with localized watermark- ing,

    R. S. Roman, P. Fernandez, H. Elsahar, A. Défossez, T. Furon, and T. Tran, “Proactive detection of voice cloning with localized watermark- ing,” inProc. Int. Conf. Mach. Learn. (ICML), Vienna, Austria, 2024, pp. 43 180–43 196

    work page 2024

  52. [52]

    Scalable watermarking for identifying large language model outputs,

    S. Dathathriet al., “Scalable watermarking for identifying large language model outputs,”Nature, vol. 634, no. 8035, pp. 818–823, 2024

    work page 2024

  53. [53]

    An invisible watermarking technique for image verification,

    M. Yeung and F. Mintzer, “An invisible watermarking technique for image verification,” inProc. Int. Conf. Image Process. (ICIP), vol. 2, Santa Barbara, CA, USA, 1997, pp. 680–683

    work page 1997

  54. [54]

    Image watermarking for tamper detection,

    J. Fridrich, “Image watermarking for tamper detection,” inProc. IEEE Int. Conf. Image Process. (ICIP), vol. 2, Chicago, IL, USA, 1998, pp. 404–408

    work page 1998

  55. [55]

    Watermarking for image authentication,

    M. Wu and B. Liu, “Watermarking for image authentication,” inProc. Int. Conf. Image Process. (ICIP), vol. 2, Chicago, IL, USA, 1998, pp. 437–441

    work page 1998

  56. [56]

    Secret and public key image watermarking schemes for image authentication and ownership verification,

    P. W. Wong and N. Memon, “Secret and public key image watermarking schemes for image authentication and ownership verification,”IEEE Trans. Image Process., vol. 10, no. 10, pp. 1593–1601, 2001

    work page 2001

  57. [57]

    Proactive deepfake defence via identity watermarking,

    Y . Zhao, B. Liu, M. Ding, B. Liu, T. Zhu, and X. Yu, “Proactive deepfake defence via identity watermarking,” inProc. IEEE/CVF Winter Conf. Appl. Comput. Vis. (WACV), Waikoloa, HI, USA, 2023, pp. 4591–4600

    work page 2023

  58. [58]

    FaceSigns: Semi-fragile watermarks for media authen- tication,

    P. Neekhara, S. Hussain, X. Zhang, K. Huang, J. McAuley, and F. Koushanfar, “FaceSigns: Semi-fragile watermarks for media authen- tication,”ACM Trans. Multimedia Comput. Commun. Appl., vol. 20, no. 11, pp. 1–21, 2024

    work page 2024

  59. [59]

    Proactive image manipulation detection via deep semi-fragile watermark,

    Y . Zhao, B. Liu, T. Zhu, M. Ding, X. Yu, and W. Zhou, “Proactive image manipulation detection via deep semi-fragile watermark,”Neuro- computing, vol. 585, pp. 1–14, 2024

    work page 2024

  60. [60]

    Image provenance analysis at scale,

    D. Moreiraet al., “Image provenance analysis at scale,”IEEE Trans. Image Process., vol. 27, no. 12, pp. 6109–6123, 2018

    work page 2018

  61. [61]

    Beyond pixels: Image provenance analysis leveraging metadata,

    A. Bharatiet al., “Beyond pixels: Image provenance analysis leveraging metadata,” inProc. IEEE Winter Conf. Appl. Comput. Vis. (WACV), Waikoloa, HI, USA, 2019, pp. 1692–1702

    work page 2019

  62. [62]

    Discovering image manipulation history by pairwise relation and forensics tools,

    X. Zhang, Z. H. Sun, S. Karaman, and S.-F. Chang, “Discovering image manipulation history by pairwise relation and forensics tools,”IEEE J. Sel. Top. Signal Process., vol. 14, no. 5, pp. 1012–1023, 2020

    work page 2020

  63. [63]

    Media forensics on social media platforms: A survey,

    C. Pasquini, I. Amerini, and G. Boato, “Media forensics on social media platforms: A survey,”EURASIP J. Inf. Secur ., pp. 1–19, 2021

    work page 2021

  64. [64]

    MFC-Prov: Media forensics challenge image provenance evaluation and data analysis on large-scale datasets,

    X. Jinet al., “MFC-Prov: Media forensics challenge image provenance evaluation and data analysis on large-scale datasets,”Neurocomputing, vol. 470, pp. 76–88, 2022

    work page 2022

  65. [65]

    SoK: Timeline based event reconstruction for digital forensics: Terminology, methodology, and current challenges,

    F. Breitinger, H. Studiawan, and C. Hargreaves, “SoK: Timeline based event reconstruction for digital forensics: Terminology, methodology, and current challenges,”F orensic Sci. Int. Digit. Investig., vol. 53, pp. 1–15, 2025

    work page 2025

  66. [66]

    Abduction, tomography, and other inverse problems,

    I. Niiniluoto, “Abduction, tomography, and other inverse problems,” Stud. Hist. Philos. Sci. A, vol. 42, no. 1, pp. 135–139, 2011

    work page 2011

  67. [67]

    The inference to the best explanation,

    G. H. Harman, “The inference to the best explanation,”Philos. Rev., vol. 74, no. 1, pp. 88–95, 1965

    work page 1965

  68. [68]

    Digital watermarking for telltale tamper proofing and authentication,

    D. Kundur and D. Hatzinakos, “Digital watermarking for telltale tamper proofing and authentication,”Proc. IEEE, vol. 87, no. 7, pp. 1167–1180, 1999

    work page 1999

  69. [69]

    Microsoft COCO: Common objects in context,

    T.-Y . Linet al., “Microsoft COCO: Common objects in context,” in Proc. Eur . Conf. Comput. Vis. (ECCV), Zurich, Switzerland, 2014, pp. 740–755

    work page 2014

  70. [70]

    U-Net: Convolutional net- works for biomedical image segmentation,

    O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional net- works for biomedical image segmentation,” inProc. Int. Conf. Med. Image Comput. Comput.-Assist. Interv. (MICCAI), Cham, 2015, pp. 234– 241

    work page 2015

  71. [71]

    Deep residual learning for image recognition,

    K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” inProc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), Las Vegas, NV , USA, 2016, pp. 770–778

    work page 2016

  72. [72]

    CBAM: Convolutional block attention module,

    S. Woo, J. Park, J.-Y . Lee, and I. S. Kweon, “CBAM: Convolutional block attention module,” inProc. Eur . Conf. Comput. Vis. (ECCV), Munich, Germany, 2018, pp. 3–19

    work page 2018

  73. [73]

    High-resolution image synthesis with latent diffusion models,

    R. Rombach, A. Blattmann, D. Lorenz, P. Esser, and B. Ommer, “High-resolution image synthesis with latent diffusion models,” inProc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), New Orleans, LA, USA, 2022, pp. 10 674–10 685

    work page 2022