Running on Zero 473 473 Florence2 + SAM2 🔥 Segment objects in images and videos using text prompts