Kuaishou Audio & Video Technology
Jun 18, 2021 · Artificial Intelligence
MiVOS: Achieving Precise Video Segmentation with Minimal User Interaction
MiVOS introduces a highly decoupled, three‑module framework—Interaction‑to‑Mask, Mask Propagation, and Difference‑aware Fusion—for interactive video object segmentation, delivering precise masks with fewer user interactions, validated on the DAVIS benchmark and supported by a new large‑scale synthetic VOS dataset (BL30K).
Computer VisionDeep Learninginteractive segmentation
0 likes · 10 min read