Mario Fritz – CISPA Helmholtz Center for Information Security

Mario Fritz – CISPA Helmholtz Center for Information Security

timely

Multi-Class Video Co-Segmentation with a Generative Multi-Video Model

We apply non-parameteric bayesian techniques to address a novel multi-class video co-segmentation challenge. Our method takes multiple videos as input and exploits commonalities between them in order to extract objects. Our formulation is based on distant-dependent Chinese Restaurant Processes.

    WordPress Theme Custom Community 2 developed by Macho Themes

    Imprint / Impressum | Data Protection / Datenschutzhinweis