Evaluation of Environmental Conditions on Object Detection using Oriented Bounding Boxes for AR Applications

TL;DR

This paper proposes a novel deep learning approach using oriented bounding boxes to enhance object detection accuracy and speed in AR, especially for small objects under varied environmental conditions.

Contribution

It introduces a detection method combining oriented bounding boxes with a deep network, evaluated on real and synthetic datasets for improved small object recognition in AR.

Findings

Better Average Precision for small objects

Improved accuracy across diverse environmental conditions

Effective detection in synthetic and real datasets

Abstract

The objective of augmented reality (AR) is to add digital content to natural images and videos to create an interactive experience between the user and the environment. Scene analysis and object recognition play a crucial role in AR, as they must be performed quickly and accurately. In this study, a new approach is proposed that involves using oriented bounding boxes with a detection and recognition deep network to improve performance and processing time. The approach is evaluated using two datasets: a real image dataset (DOTA dataset) commonly used for computer vision tasks, and a synthetic dataset that simulates different environmental, lighting, and acquisition conditions. The focus of the evaluation is on small objects, which are difficult to detect and recognise. The results indicate that the proposed approach tends to produce better Average Precision and greater accuracy for small objects in most of the tested conditions.