You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
A temporal computer vision system for detecting missed polyps in colonoscopies using a Dual-Path architecture (Spatial RetinaNet + Temporal ConvGRU) and a clinical fusion engine to reduce false positives.
This repository is a collection of Python scripts and Jupyter notebooks for understanding the performance improvement in image classification, object detection and instance segmentation with OpenVINO. It also contains reference implementations of dwell time analytics, ALPR and polyp detection.
This repository contains the BKM for training YOLOv11n model on Intel Arc A770 GPU and the reference implementation of polyp detection in colonoscopy video with the optimized model using OpenVINO 2025
This research will show an innovative method useful in the segmentation of polyps during the screening phases of colonoscopies. To do this we have adopted a new approach which consists in merging the hybrid semantic network (HSNet) architecture model with the Reagion-wise(RW) as a loss function for the backpropagation process.
Official repo of "EndoBoost: a plug-and-play module for false positive suppression during computer-aided polyp detection in real-world colonoscopy (with dataset)"