Media companies have accumulated vast digital archives and collections of images and videos over the years. Since these collections have been gradually and iteratively built over time, often by different departments and units of media companies, they usually have little or no metadata such as tags, categories, and other types of annotations. This lack of coherent media asset organisation tailored to the media company business and services precludes the successful monetisation of these media assets and the creation and offering of new services. In addition, both big traditional media companies and more so digital media platforms combine in their collections both media content, created by these companies, but increasingly also user-generated content (UGC). Such hybrid media archives need advanced content moderation (CM) solutions, often working in real-time to safeguard viewers and meet laws and regulations.
To address these challenges, this use case utilizes several AI-enabled tools such as visual tagging, categorization, and content moderation to facilitate:
- a) Automated (re)organisation of large media collections of photos and video: Techniques for mapping existing taxonomies and ontologies used by media companies, to restructure them to more optimal ones, and training of specialized models for such companies using state-of-the-art CNNs for media asset categorisation will be used. The know-how of consortium partners with personal and enterprise photo collection organisation will be extended to mixed collections of photos and video.
- b) Automated and human-in-the-loop moderation of user-generated media content: The use case will also make use of existing CM solutions that combine automated detection of diverse and customizable types of inappropriate content, such as weapons, drugs, nudity, pornography, etc. with a CM platform and option to add internal or external data teams to verify inappropriate content flagged by AI algorithms. As part of this use case, Imagga’s CM platform will be tested and further optimised with mixed and hybrid media collections and platforms where UGC can come both as stand-alone new visual content (photos, video) or as comments to existing media assets, such as films, news, etc. Additional focus will be on implementing a live-stream optimization module for optimizing costs and precision of automated real-time moderation. This will meet the growing demand of media companies to have scalable real-time content moderation to safeguard users and meet various regulations.
Additionally, the two demonstrators presented in this use case will be tested by target groups of real users from media companies in Living Lab trials, in order to validate their performance and help tailor them further to meet user requirements.
Immaga’s Content Moderation platform will be enriched with AI functionalities delivered by AI4Media, including automatic filtering of unsafe content in images, videos and live streams.