Detection of Splicing in Digital Image Forgery

D 2026

Detection of Splicing in Digital Image Forgery

PITSMAUSOVÁ, Zuzana a David SVOBODA

Základní údaje

Originální název

Detection of Splicing in Digital Image Forgery

Autoři

PITSMAUSOVÁ, Zuzana a David SVOBODA

Vydání

445 Hoes Lane Piscataway, NJ 08854, USA, Proceedings on 2026 IEEE International Conference on Image Processing (ICIP), od s. 1-6, 6 s. 2026

Nakladatel

IEEE

Další údaje

Jazyk

angličtina

Typ výsledku

Stať ve sborníku

Obor

10200 1.2 Computer and information sciences

Stát vydavatele

Spojené státy

Utajení

není předmětem státního či obchodního tajemství

Forma vydání

elektronická verze "online"

Odkazy

URL

Označené pro přenos do RIV

Ano

Organizační jednotka

Fakulta informatiky

ISBN

979-8-3315-5151-3

ISSN

DOI

https://doi.org/10.1109/ICIP61757.2026.11630551

Klíčová slova anglicky

Splicing; Location awareness; Modeling; Printing; Forensics; Training; Signal detection; Deep learning; Conferences; Forgery; Image Forensics; Image Splicing Localization; Deep Learning; Transformer-Based Architecture

Štítky

cbia-web

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 5. 10. 2026 09:45, doc. RNDr. David Svoboda, Ph.D.

Anotace

V originále

Image splicing is a challenging form of digital manipulation in which content from multiple sources is combined into a single image, often leaving subtle forensic traces. Prior work has increasingly highlighted that splicing localization differs from standard semantic segmentation in its objectives and the nature of the cues it relies on. Building upon these insights, this paper presents a lightweight hierarchical transformer-based framework for pixel-level image splicing localization, in which both the architecture and training strategy are deliberately aligned with the forensic characteristics of the task. The proposed approach emphasizes resolution preservation, multi-scale supervision, and boundary sensitivity to better capture subtle manipulation traces while limiting dominance of semantic image content. Unlike heavier multi-stream solutions, the method adopts a compact single-stream design focused on forensic signals. The approach is broadly evaluated using quantitative and qualitative analysis with a focus on splicing forgeries, demonstrating stable localization behavior across diverse datasets.

Návaznosti

MUNI/A/1784/2025, interní kód MU

Název: Modelování, analýza a verifikace (2026)

Investor: Masarykova univerzita, Modelování, analýza a verifikace (2026)

VK01010147, projekt VaV

Název: Automatizovaná forenzní laboratoř digitálních dat pro odhalování komplexní trestné činnosti

Investor: Ministerstvo vnitra ČR, Automatizovaná forenzní laboratoř digitálních dat pro odhalování komplexní trestné činnosti

Citovat

PITSMAUSOVÁ, Zuzana a David SVOBODA. Detection of Splicing in Digital Image Forgery. Online. In Proceedings on 2026 IEEE International Conference on Image Processing (ICIP). 445 Hoes Lane Piscataway, NJ 08854, USA: IEEE, 2026, s. 1-6. ISBN 979-8-3315-5151-3. Dostupné z: https://doi.org/10.1109/ICIP61757.2026.11630551.

@inproceedings{2591518,
   author = {Pitsmausová, Zuzana and Svoboda, David},
   address = {445 Hoes Lane Piscataway, NJ 08854, USA},
   booktitle = {Proceedings on 2026 IEEE International Conference on Image Processing (ICIP)},
   doi = {https://doi.org/10.1109/ICIP61757.2026.11630551},
   keywords = {Splicing; Location awareness; Modeling; Printing; Forensics; Training; Signal detection; Deep learning; Conferences; Forgery; Image Forensics; Image Splicing Localization; Deep Learning; Transformer-Based Architecture},
   howpublished = {elektronická verze "online"},
   language = {eng},
   location = {445 Hoes Lane Piscataway, NJ 08854, USA},
   isbn = {979-8-3315-5151-3},
   pages = {1-6},
   publisher = {IEEE},
   title = {Detection of Splicing in Digital Image Forgery},
   url = {https://ieeexplore.ieee.org/document/11630551},
   year = {2026}
}

TY  - CONF
ID  - 2591518
AU  - Pitsmausová, Zuzana - Svoboda, David
PY  - 2026
TI  - Detection of Splicing in Digital Image Forgery
PB  - IEEE
CY  - 445 Hoes Lane Piscataway, NJ 08854, USA
SN  - 9798331551513
KW  - Splicing
KW  - Location awareness
KW  - Modeling
KW  - Printing
KW  - Forensics
KW  - Training
KW  - Signal detection
KW  - Deep learning
KW  - Conferences
KW  - Forgery
KW  - Image Forensics
KW  - Image Splicing Localization
KW  - Deep Learning
KW  - Transformer-Based Architecture
UR  - https://ieeexplore.ieee.org/document/11630551
N2  - Image splicing is a challenging form of digital manipulation in which content from multiple sources is combined into a single image, often leaving subtle forensic traces. Prior work has increasingly highlighted that splicing localization differs from standard semantic segmentation in its objectives and the nature of the cues it relies on. Building upon these insights, this paper presents a lightweight hierarchical transformer-based framework for pixel-level image splicing localization, in which both the architecture and training strategy are deliberately aligned with the forensic characteristics of the task. The proposed approach emphasizes resolution preservation, multi-scale supervision, and boundary sensitivity to better capture subtle manipulation traces while limiting dominance of semantic image content. Unlike heavier multi-stream solutions, the method adopts a compact single-stream design focused on forensic signals. The approach is broadly evaluated using quantitative and qualitative analysis with a focus on splicing forgeries, demonstrating stable localization behavior across diverse datasets.
ER  -

PITSMAUSOVÁ, Zuzana a David SVOBODA. Detection of Splicing in Digital Image Forgery. Online. In \textit{Proceedings on 2026 IEEE International Conference on Image Processing (ICIP)}. 445 Hoes Lane Piscataway, NJ 08854, USA: IEEE, 2026, s.~1-6. ISBN~979-8-3315-5151-3. Dostupné z: https://doi.org/10.1109/ICIP61757.2026.11630551.

Přehled o publikaci