The Uncovr project aims to use a combination of artificial and collective intelligence to identify musical works in videos, including cover versions and background music, in order to ensure fair compensation for artists. This involves linking existing web resources with metadata and using collective intelligence through serious games or crowdsourcing to supplement and consolidate the information. The project will explore the use of machine learning methods, in particular deep neural networks, to enable automatic identification and linking. Challenges include the selection of appropriate datasets and the development of methods that intelligently combine algorithmic and human recognition performance.