Fetching the paper…

BAVS: Bootstrapping Audio-Visual Segmentation by Integrating Foundation Knowledge · Around