Fail-Safe Human Detection for Drones Using a Multi-Modal Curriculum Learning Approach

TL;DR

This paper introduces a multi-modal fusion approach using a new dataset and curriculum learning to improve drone safety-critical human detection, especially under challenging conditions, by combining cameras and radar data.

Contribution

It presents KUL-UAVSAFE, a novel dataset for drone safety, and proposes SAUL, a curriculum learning strategy that enhances multi-modal fusion robustness for human detection.

Findings

15% improvement in peak F1 score over previous methods

Robust real-time detection demonstrated on edge computing hardware

Effective fusion of cameras and radar for drone safety applications

Abstract

Drones are currently being explored for safety-critical applications where human agents are expected to evolve in their vicinity. In such applications, robust people avoidance must be provided by fusing a number of sensing modalities in order to avoid collisions. Currently however, people detection systems used on drones are solely based on standard cameras besides an emerging number of works discussing the fusion of imaging and event-based cameras. On the other hand, radar-based systems provide up-most robustness towards environmental conditions but do not provide complete information on their own and have mainly been investigated in automotive contexts, not for drones. In order to enable the fusion of radars with both event-based and standard cameras, we present KUL-UAVSAFE, a first-of-its-kind dataset for the study of safety-critical people detection by drones. In addition, we propose a baseline CNN architecture with cross-fusion highways and introduce a curriculum learning strategy for multi-modal data termed SAUL, which greatly enhances the robustness of the system towards hard RGB failures and provides a significant gain of 15% in peak F1 score compared to the use of BlackIn, previously proposed for cross-fusion networks. We demonstrate the real-time performance and feasibility of the approach by implementing the system in an edge-computing unit. We release our dataset and additional material in the project home page.