PRISM: A Promptable and Robust Interactive Segmentation Model with Visual Prompts

arXiv:2404.15028v1 Announce Type: new
Abstract: In this paper, we present PRISM, a Promptable and Robust Interactive Segmentation Model, aiming for precise segmentation of 3D medical images. PRISM accepts various visual inputs, including points, boxes, and scribbles as sparse prompts, as well as masks as dense prompts. Specifically, PRISM is designed with four principles to achieve robustness: (1) Iterative learning. The model produces segmentations by using visual prompts from previous iterations to achieve progressive improvement. (2) Confidence learning. PRISM employs multiple …

arxiv cs.cv interactive prompts robust segmentation type visual