Archangel: A Hybrid UAV-based Human Detection Benchmark with Position and Pose Metadata

TL;DR

Archangel is a comprehensive UAV-based human detection dataset with detailed position and pose metadata, enabling improved model evaluation and understanding of UAV imagery variations.

Contribution

It introduces the first UAV-based object detection dataset with synchronized real and synthetic data and rich metadata for enhanced model diagnosis and learning.

Findings

Metadata improves model evaluation accuracy

Synthetic data complements real data effectively

Insights into model optimization with metadata

Abstract

Learning to detect objects, such as humans, in imagery captured by an unmanned aerial vehicle (UAV) usually suffers from tremendous variations caused by the UAV's position towards the objects. In addition, existing UAV-based benchmark datasets do not provide adequate dataset metadata, which is essential for precise model diagnosis and learning features invariant to those variations. In this paper, we introduce Archangel, the first UAV-based object detection dataset composed of real and synthetic subsets captured with similar imagining conditions and UAV position and object pose metadata. A series of experiments are carefully designed with a state-of-the-art object detector to demonstrate the benefits of leveraging the metadata during model evaluation. Moreover, several crucial insights involving both real and synthetic data during model optimization are presented. In the end, we discuss the advantages, limitations, and future directions regarding Archangel to highlight its distinct value for the broader machine learning community.