Two independent ML/CV researchers (M.Eng, ex-research-institute in Europe) looking for an arXiv cs.CV endorser for a nearly finished paper. Happy to share the full draft, repo, or talk collaboration [D]
Two independent researchers are seeking an endorsement for their paper on a new computer vision system called Locate-SAM2. This system connects NVIDIA's LocateAnything-3B with Meta's SAM 2.1 through a lightweight adapter, aiming to determine if the choice of grounder impacts mask quality in a modular text-to-mask pipeline. Their work demonstrates competitive performance on the RefCOCO dataset, achieving 0.772 mIoU, and includes detailed comparisons, ablation studies, and analysis of failure cases. AI
IMPACT Researchers are seeking community validation for a new computer vision system that integrates existing models to improve mask generation.