Junlong Cheng, Jin Ye, Zhongying Deng, Jianpin Chen, Tianbin Li...
TL;DRSAM-Med2D 是目前最全面的研究,通过收集和整理公开和私有数据集的约 4.6M 图像和 19.7M 掩膜构建了一个包括不同模态和对象的大规模医学图像分割数据集,并通过包围盒、点和掩膜的综合提示将自然图像分割模型 SAM 应用于医学图像分割,进行了彻底的微调,获得了最佳性能和泛化能力。
Abstract
The segment anything model (SAM) represents a state-of-the-art research
advancement in natural image segmentation, achieving impressive results with
input prompts such as points and bounding boxes. However, our evaluation and
recent research indicate that directly applying the pretrain