Abstract
Video and music clips in MTV match together in particular ways to produce attractive effect. In this paper, we use a dual-wing harmonium model to learn and represent the underlying association patterns between music and video clips in professional MTV. We also use the discovered patterns to facilitate automatic MTV generation. Provided with a raw video and certain professional MTV as template, we generate a new MTV by efficiently inferring the most related video clip for every music clip based on the trained model. Our method shows encouraging result compared with other automatic MTV generation approach.Keywordsmusic video generationharmonium modelassociation pattern
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have