Advances in AI-Generated Images and Videos.
In recent years generative AI models and tools have experienced a significant increase, especially techniques to generate synthetic multimedia content, such as images or videos. These methodologies present a wide range of possibilities; however, they can also present several risks that should be taken into account. In this survey we describe in detail different techniques for generating synthetic multimedia content, and we also analyse the most recent techniques for their detection. In order to achieve these objectives, a key aspect is the availability of datasets, so we have also described the main datasets available in the state of the art. Finally, from our analysis we have extracted the main trends for the future, such as transparency and interpretability, the generation of multimodal multimedia content, the robustness of models and the increased use of diffusion models. We find a roadmap of deep challenges, including temporal consistency, computation requirements, generalizability, ethical aspects, and constant adaptation.
Authors
- Abdenour Hadid (ORCID: https://orcid.org/0000-0001-9092-735X)
- Helena Liz (ORCID: https://orcid.org/0000-0003-4962-6314)
- David Camacho (ORCID: https://orcid.org/0000-0002-5051-3475)
- Mamadou Keïta (ORCID: https://orcid.org/0009-0009-7618-9253)
- Wassim Hamidouche (ORCID: https://orcid.org/0000-0002-0143-1756)
- Abdelmalik Taleb‐Ahmed (ORCID: https://orcid.org/0000-0001-7218-3799)
- A. Miguel San Martin
- Hessen Bougueffa
Institutions
- Centre National de la Recherche Scientifique (FR)
- Université de Lille (FR)
- Sorbonne Université (FR)
- Institut d'Électronique et des Technologies du numéRique (FR)
- Institut d'Electronique, de Microélectronique et de Nanotechnologie (FR)
- Sorbonne University Abu Dhabi (AE)
- Université de Rennes (FR)
- Université Polytechnique Hauts-de-France (FR)
- École Centrale de Lille (FR)
- Universidad Politécnica de Madrid (ES)
Publication Details
- Journal
- International Journal of Interactive Multimedia and Artificial Intelligence
- Published
- 2024-11-29
- DOI
- https://doi.org/10.9781/ijimai.2024.11.003
- Citations
- 5
- Primary Topic
- Generative Adversarial Networks and Image Synthesis
- Type
- article
- Field-Weighted Citation Impact
- 1.25
Funders
- European Commission
- Total
- Sorbonne Université
- Agencia Estatal de Investigación