Sa2VA: Marrying SAM2 with LLaVA for Dense Grounded Understanding of Images and Videos Paper • 2501.04001 • Published about 1 month ago • 42
