AOD-YOLOv8: arbitrary-oriented object detection using CRIoU loss function

Detecting objects with specific orientations in aerial imagery is a complex task due to the random rotations and high density of objects within the images. Additionally, boundary discontinuity from regression parametrization poses a major challenge in designing loss functions for rotated object detection. Several bounding box representations and their associated loss functions are being gradually developed to address this challenge. In this study, we compare loss functions for oriented bounding boxes including Probabilistic IoU (ProbIoU), KFIoU and others. The evaluation uses a large scale aerial benchmark dataset DOTAv1.0, which includes 15 categories and DOTA-RBD, a subset of DOTAv1.0, for evaluation on square objects. We also introduce Complete Rotated IoU (CRIoU) loss which is improved over Rotated IoU (RIoU) loss. CRIoU and ProbIoU loss achieved mAP score of 64.3% and 63.4% on the test set while KFIoU and RIoU losses got 63.3% and 62.2% respectively. CRIoU demonstrates marginal improvement in oriented bounding box localization, supported by the brief experimental and visual analysis.