Long Range Object-Level Monocular Depth Estimation for UAVs

TL;DR

This paper enhances monocular object detection for UAVs by integrating depth estimation as a regression and classification task, improving long-range detection accuracy crucial for autonomous collision avoidance.

Contribution

It introduces novel depth encoding methods and a Soft-Argmax loss function within YOLOX, advancing monocular depth estimation for UAV object detection.

Findings

Proposed methods outperform existing state-of-the-art approaches.

Introduced the Fitness score metric for joint detection and depth evaluation.

Achieved improved long-range detection accuracy on UAV datasets.

Abstract

Computer vision-based object detection is a key modality for advanced Detect-And-Avoid systems that allow for autonomous flight missions of UAVs. While standard object detection frameworks do not predict the actual depth of an object, this information is crucial to avoid collisions. In this paper, we propose several novel extensions to state-of-the-art methods for monocular object detection from images at long range. Firstly, we propose Sigmoid and ReLU-like encodings when modeling depth estimation as a regression task. Secondly, we frame the depth estimation as a classification problem and introduce a Soft-Argmax function in the calculation of the training loss. The extensions are exemplarily applied to the YOLOX object detection framework. We evaluate the performance using the Amazon Airborne Object Tracking dataset. In addition, we introduce the Fitness score as a new metric that jointly assesses both object detection and depth estimation performance. Our results show that the proposed methods outperform state-of-the-art approaches w.r.t. existing, as well as the proposed metrics.