Polyp SAM 2: Advancing Zero-Shot Polyp Segmentation in Colorectal Cancer Detection

The segmentation of polyps is crucial to the early detection and diagnosis of colorectal cancer. However, obtaining accurate segmentation often requires labor-intensive annotations and specialized models. Lately, Meta AI Research released the Segment Anything Model 2 (SAM 2), which provides bright results in a number of segmentation assignments. In this report, we analyze the capabilities of SAM 2 in segmenting polyps under various prompted conditions. We expect that this report will contribute to the improvement of polyp segmentation and stimulate further research into this field. This project is publicly available at https://github.com/sajjad-sh33/Polyp-SAM-2.