Official Detectron2 implementation of DA-RetinaNet of our Image and Vision Computing 2021 work 'An unsupervised domain adaptation scheme for single-stage artwork recognition in cultural sites'

Detectron2 implementation of DA-Faster R-CNN, Domain Adaptive Faster R-CNN for Object Detection in the Wild

Official Detectron2 implementation of STMDA-RetinaNet of our Computer Vision and Image Understanding 2022 work 'A Multi Camera Unsupervised Domain Adaptation Pipeline for Object Detection in Cultural Sites through Adversarial Learning and Self-Training'

Fast alternative to FiftyOne for creating a subset of the COCO dataset.

FROC analysis for COCO detections for Detectron(2) and OpenMMLab

AnnotateMate: Open-source tool to analyse, modify COCO format annotation

Python script generates colored masks from COCO-style annotations. It reads the COCO annotation files, creates masks for each annotation, colors the masks based on the annotation's category, and saves the colored masks as images. The script is designed to handle multiple categories and can be easily extended to support more categories if needed.

Segmentation tool for converting the AI segmentation model inference back into the coco annotator.