YUDO: YOLO for Uniform Directed Object Detection

TL;DR

This paper introduces YUDO, a specialized, efficient object detection model for uniform directed objects that predicts position and orientation without size estimation, utilizing a novel rotated IoU metric and minimal detection heads.

Contribution

The work adapts YOLO architecture for directed object detection, introduces DirIoU for rotated boxes, and demonstrates the effectiveness of a simplified detection head setup.

Findings

YUDO achieves accurate directed object detection with minimal model complexity.

The extended DirIoU improves matching and NMS for rotated bounding boxes.

A tiny YOLO variant performs well on the Honeybee dataset.

Abstract

This paper presents an efficient way of detecting directed objects by predicting their center coordinates and direction angle. Since the objects are of uniform size, the proposed model works without predicting the object's width and height. The dataset used for this problem is presented in Honeybee Segmentation and Tracking Datasets project. One of the contributions of this work is an examination of the ability of the standard real-time object detection architecture like YoloV7 to be customized for position and direction detection. A very efficient, tiny version of the architecture is used in this approach. Moreover, only one of three detection heads without anchors is sufficient for this task. We also introduce the extended Skew Intersection over Union (SkewIoU) calculation for rotated boxes - directed IoU (DirIoU), which includes an absolute angle difference. DirIoU is used both in the matching procedure of target and predicted bounding boxes for mAP calculation, and in the NMS filtering procedure. The code and models are available at https://github.com/djordjened92/yudo.