Overview
The Media Asset Annotation Management (MAAM) platform is a web-based tool designed to support media professionals, journalists, and fact-checkers in verifying the authenticity of visual content. Built to address the growing threat of visual disinformation, MAAM combines advanced media asset management capabilities with state-of-the-art AI analysis for detecting manipulation in images and videos.
MAAM provides users with a streamlined interface for uploading, organizing, annotating, and searching through large collections of multimedia content. Through dedicated workspaces and filtering options, users can efficiently manage their assets in collaborative environments.
But MAAM doesn’t stop at media management — it also empowers users with AI-powered disinformation detection. The platform integrates a suite of cutting-edge algorithms capable of identifying a wide range of manipulation techniques, including deepfakes, synthetic media generation, and image tampering. For each analysis, MAAM provides confidence scores, visual overlays, and human-readable explanations, helping users interpret the results even without technical expertise.
Additional features such as geolocation estimation, visual similarity search, and reverse image/video lookup allow users to investigate media origins and context more efficient.
Key People
Publications
-
(2023). MAAM: Media Asset Annotation and Management. Proceedings of the 2023 ACM International Conference on Multimedia Retrieval.
