Aims & Scope

The central focus of this journal is the computer analysis of pictorial information. Computer Vision and Image Understanding publishes papers covering all aspects of image analysis from the low-level, iconic processes of early vision to the high-level, symbolic processes of recognition and interpretation. A wide range of topics in the image understanding area is covered, including papers offering insights that differ from predominant views.

Research Areas Include:

    Theory
    Early vision
    Data structures and representations
    Shape
    Range
    Motion
    Matching and recognition
    Architecture and languages
    Vision systems

Special Issue on Multimedia Forensics in Practice: Detecting and Interpreting Synthetic Media (MF-DISM)
截稿日期: 2026-03-31

Generative models and accessible editing tools have accelerated the production of synthetic and manipulated images, videos, and audio, undermining trust in digital evidence and media-based decision systems. Manipulations affect authentication, journalism, law enforcement, and biometric security, where undetected forgeries enable identity fraud, misinformation, reputational damage, and procedural errors. The Special Issue Multimedia Forensics in Practice: Detecting and Interpreting Synthetic Media (MF-DISM) sets a research agenda for multimedia forensics that is simultaneously rigorous and deployable, unifying detection, localisation, and source attribution with interpretable, calibrated outputs that remain reliable under compression, editing, scaling, and recapture. Scope spans cross-modal pipelines as text-to-image, text-to-video, and voice cloning, and the forensic implications of diffusion models and LLM-guided synthesis, with emphasis on provenance, model and camera fingerprinting, and audio–video consistency. Methodological expectations include verified generalisation across datasets and attack families, adversarial robustness, stress testing, strong baselines, ablations, and reproducible datasets tailored to evolving threats. Application domains cover biometric authentication, content verification, platform integrity, and investigative workflows. The Special Issue’s role is to consolidate standards of evidence, provide benchmarks that reflect operational conditions, and translate advances in computer vision, signal processing, and cybersecurity into accountable tools that institutions can adopt at scale.

The diffusion of generative models and editing tools has sharply increased the creation and spread of synthetic and manipulated content across images, video, and audio. Such material threatens the reliability of digital evidence and the integrity of media-based systems for authentication, communication, and decision making. In applied domains (law enforcement, biometric security, journalism, and real-time verification) undetected forgeries can enable identity fraud, misinformation, reputational damage, judicial errors, and the circumvention of biometric systems. Addressing these risks requires forensic solutions that combine high accuracy with interpretability and operational viability.

The Special Issue MF-DISM advances multimedia forensics that is scientifically rigorous and deployable at scale. Focus lies on detection, localisation, and source attribution with calibrated outputs that remain reliable under compression, scaling, editing, and recapture. Core computer-vision challenges motivate the call: robust visual analysis, manipulation detection in image and video, and principled modelling of temporal dynamics and spatio-temporal correlations.

Scope covers cross-modal pipelines (audio, video, image), including text-to-image, text-to-video, and voice cloning, and examines diffusion-based generation and LLM-guided synthesis with emphasis on cross-modal consistency, model attribution, and manipulation provenance. Forensics-driven synthetic and adversarial datasets are encouraged to support reproducible benchmarking. Equal priority goes to robustness, explainability, adversarial resilience, and scalable deployment for platform integrity and investigative workflows.

Submissions at the intersection of artificial intelligence, signal processing, and cybersecurity are welcome. Topics include, but are not limited to:

    Detection, localisation, and attribution of manipulated media (deepfakes, face swaps, style transfer, morphing, audio spoofing)
    Generalisable forgery detection with transfer across datasets, codecs, and attack families
    Cross-modal analysis (audio–video consistency; validation of text-to-image/video content)
    Explainable, trustworthy systems with calibrated uncertainty
    Multimedia forensics in biometric applications (face, fingerprint, voice)
    Creation and validation of synthetic or adversarial datasets reflecting evolving threats
    Attribution and fingerprinting of generative models and camera pipelines
    Adversarial robustness for vision-based manipulation detection
    Secure, scalable architectures and real-time operation
    Deployment studies and evaluations in operational environments
    Ethics, governance, and responsible use of synthetic media

Overall objective: strengthen trust in multimedia content through auditable, reproducible methods and resources ready for adoption at scale in complex, adversarial, and multimodal settings.

Guest editors:
Dr. Marco Micheletto
Università degli Studi di Cagliari, Cagliari, Italy
Dr. Marta Gomez-Barrero
University of the Bundeswehr Munich, Neubiberg, Germany
Dr. Luca Guarnera
Università degli Studi di Catania, Catania, Italy
Dr. Giulia Orrù
Università degli Studi di Cagliari, Cagliari, Italy
Dr. Enjie Ghorbel
University of Luxembourg, Esch-sur-Alzette, Luxembourg

Manuscript submission information:

    Open for Submission: from 01-Dec-2025 to 31-Mar-2026
    Submission Site: Editorial Manager®
    Article Type Name: "VSI: YCVIU_MF-DISM" - please select this item when you submit manuscripts online

All manuscripts will be peer-reviewed. Submissions will be evaluated based on originality, significance, technical quality, and clarity. Once accepted, articles will be posted online immediately and published in a journal regular issue within weeks. Articles will also be simultaneously collected in the online special issue.
For any inquiries about the appropriateness of contribution topics, welcome to contact Leading Guest Editor (Dr. Marco Micheletto).
Guide for Authors will be helpful for your future contributions, read more: Guide for authors - Computer Vision and Image Understanding - ISSN 1077-3142 | ScienceDirect.com by Elsevier
For more information about our Journal, please visit our ScienceDirect Page: CVIU | Computer Vision and Image Understanding | Journal | ScienceDirect.com by Elsevier

Keywords:

    multimedia forensics
    deepfake detection
    source attribution
    model fingerprinting
    explainable AI
    presentation attack detection
    biometric security
    adversarial robustness
    diffusion models
    synthetic datasets