Digital Asset Management using artificial intelligence – DAMIALabs

Since the 2000s, digital images and videos have revolutionized the way communication and marketing work. All sectors of activity have progressively had to address the issue of the distribution and preservation of their digital media. Commerce, brands, services or culture have progressively switched their communications to digital. All of these actors quickly needed a tool dedicated to the management of digital media: "Digital Asset Management". The "digital assets" refer to all the contents of the photo, audio files, videos, plans or composite documents (packaging, catalog, PDF, etc.). DAM" solutions first integrated the know-how of documentalists and archivists. Then, the solutions evolved towards project management optimizing the use of assets. Today, the life cycle of media content has greatly accelerated with multi-channel circuits: e-commerce, catalogs, social networks, etc.

The joint laboratory (LabCom DAMIALabs) proposed by the CNRS XLIM institute and the company Einden is positioned in this context of DAM solutions, with the central issue of indexing and searching media for large databases of thematic images or associated with a documentary collection. The scientific and technical challenge addressed concerns the deployment of innovative solutions that respond in a complementary way to the problem of content-based indexing. More precisely, we propose to design and deploy methods coming from Artificial Intelligence (AI) and data mining, allowing to characterize, organize and index media bases. Methodologically, we propose to rely on approaches based on "raw" images or on modeling (with a competition) and to exploit machine learning methods by integrating the expert in the loop. This question of the integration of the expert, of Active Learning, but also of the construction of feedback loops following reinforcement learning, is one of the scientific challenges addressed in this LabCom.