QueenVIS: Rethinking Image-Only Training for Video Instance Segmentation via Query Enrichment Paper • 2607.24598 • Published Jul 27 • 9