Fig. 2
From: A fine-tuned foundational model SurgiSAM2 for surgical video anatomy segmentation and detection
The number of annotated masks (x-axis) for the five datasets (CholecSeg8k, Dresden, Endoscapes, UreterUD, and m2caiSeg), categorized by training, validation, and testing splits across all organ/tissue classes (y-axis). The CholecSeg8k dataset constituted the largest portion across all dataset splits.
