From Generative Adversarial Networks (GANs) to Diffusion Models, the realm of AI content production is evolving at an unprecedented pace. The rapid progression of these technologies poses both opportunities and challenges. On the one hand, they enable the creation of remarkable content for entertainment and artistic expression. On the other, these technologies can be used to manipulate information and spread disinformation. The general public is flooded with these contents via social media platforms and various news agencies, making it increasingly difficult to distinguish facts from fiction. For these reasons, the research branch called multimedia forensics exploits imperceptible traces left by generative algorithms capable of retrospectively detecting whether a multimedia content has been generated or not by artificial intelligence systems. These algorithms can play a crucial role in the current information landscape by detecting fake media and, ultimately, reinforcing trust in digital information.
In the ever-evolving digital landscape, multimedia forensics has emerged as a crucial discipline to ensure multimedia information's authenticity, trustworthiness, and accuracy. This research topic brings together researchers and experts in these fields to tackle pressing challenges related to identifying the source and trustworthiness of both human-generated and AI-produced content.
Firstly, we aim to explore techniques that enable us to distinguish between genuine and synthetic image and video data. This includes recognizing traces of generative models within media content and developing methods for determining their authenticity.
Secondly, we intend to examine tools and strategies that enable news agencies and content providers to verify the veracity and provenance of their information in the face of increasingly sophisticated AI manipulation techniques.
Lastly, by encouraging the creation of comprehensive datasets consisting of both AI-generated and genuine multimedia contents, this research topic aims to provide a solid base for training and evaluating new tools and technologies for media forensics and digital content verification.
The scope of this research topic encompasses algorithms and technologies capable of detecting synthetic images and videos generated by AI models, along with data that can be used to train and evaluate such algorithms. We accept papers in the form of Original Research, Methods, Systematic Review, Hypothesis & Theory, Technology and Code, Mini Review, and Data Report that contribute to the following topics but not limited to:
● AI techniques for multimedia forensics
● Identification of synthetic visual content generated by AI algorithms
● Algorithms for detecting manipulated media spread across social networks, web and instant messaging platforms
● Methods to trace the provenance of image and video data
● Tools and applications designed for end-users to detect AI-generated media
Keywords:
Multimedia Forensics, Artificial Intelligence, Text-to-Image Detection, Synthetic Images and Videos, DeepFakes
Important Note:
All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
From Generative Adversarial Networks (GANs) to Diffusion Models, the realm of AI content production is evolving at an unprecedented pace. The rapid progression of these technologies poses both opportunities and challenges. On the one hand, they enable the creation of remarkable content for entertainment and artistic expression. On the other, these technologies can be used to manipulate information and spread disinformation. The general public is flooded with these contents via social media platforms and various news agencies, making it increasingly difficult to distinguish facts from fiction. For these reasons, the research branch called multimedia forensics exploits imperceptible traces left by generative algorithms capable of retrospectively detecting whether a multimedia content has been generated or not by artificial intelligence systems. These algorithms can play a crucial role in the current information landscape by detecting fake media and, ultimately, reinforcing trust in digital information.
In the ever-evolving digital landscape, multimedia forensics has emerged as a crucial discipline to ensure multimedia information's authenticity, trustworthiness, and accuracy. This research topic brings together researchers and experts in these fields to tackle pressing challenges related to identifying the source and trustworthiness of both human-generated and AI-produced content.
Firstly, we aim to explore techniques that enable us to distinguish between genuine and synthetic image and video data. This includes recognizing traces of generative models within media content and developing methods for determining their authenticity.
Secondly, we intend to examine tools and strategies that enable news agencies and content providers to verify the veracity and provenance of their information in the face of increasingly sophisticated AI manipulation techniques.
Lastly, by encouraging the creation of comprehensive datasets consisting of both AI-generated and genuine multimedia contents, this research topic aims to provide a solid base for training and evaluating new tools and technologies for media forensics and digital content verification.
The scope of this research topic encompasses algorithms and technologies capable of detecting synthetic images and videos generated by AI models, along with data that can be used to train and evaluate such algorithms. We accept papers in the form of Original Research, Methods, Systematic Review, Hypothesis & Theory, Technology and Code, Mini Review, and Data Report that contribute to the following topics but not limited to:
● AI techniques for multimedia forensics
● Identification of synthetic visual content generated by AI algorithms
● Algorithms for detecting manipulated media spread across social networks, web and instant messaging platforms
● Methods to trace the provenance of image and video data
● Tools and applications designed for end-users to detect AI-generated media
Keywords:
Multimedia Forensics, Artificial Intelligence, Text-to-Image Detection, Synthetic Images and Videos, DeepFakes
Important Note:
All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.