Researchers have developed ViewSAM, a novel framework for weakly supervised Cross-view Referring Multi-Object Tracking (CRMOT). This approach leverages foundation models like SAM2 and SAM3 to generate pseudo-supervision, reducing the need for costly frame-level annotations. ViewSAM explicitly models view-aware cross-modal semantics, enabling robust tracking across different camera perspectives with minimal additional parameters. AI
Summary written by gemini-2.5-flash-lite from 2 sources. How we write summaries →
IMPACT Introduces a more efficient method for multi-object tracking across camera views by reducing reliance on extensive annotations.
RANK_REASON The cluster contains a research paper detailing a new model and framework for a specific computer vision task.