
TransRUPNet, a transformer-based encoder-decoder architecture, achieves accurate and real-time polyp segmentation with strong generalization capabilities across diverse colonoscopy datasets.


coremsg

transrupnet-a-transformer-based-architecture-for-accurate-and-real-time-polyp-segmentation-in-colonoscopy


TransRUPNet: A Transformer-based Architecture for Accurate and Real-Time Polyp Segmentation in Colonoscopy


title_rewrite


A novel self-prompting polyp segmentation model that integrates the YOLOv8 object detection and SAM 2 segmentation models to achieve high accuracy and efficiency with reduced annotation effort.


autonomous-polyp-segmentation-in-colonoscopy-using-a-hybrid-yolo-sam-2-model


Autonomous Polyp Segmentation in Colonoscopy using a Hybrid YOLO-SAM 2 Model
