LER-YOLO: Reliability-Aware Expert Routing for Misaligned RGB-Infrared UAV Detection

TL;DR

LER-YOLO is a novel detection framework that enhances RGB-infrared UAV detection by incorporating reliability-aware expert routing to handle sensor misalignment and improve fusion accuracy.

Contribution

It introduces a reliability-aware sparse mixture-of-experts framework with an uncertainty-aware alignment and adaptive expert selection for better cross-modal fusion.

Findings

Achieves 89.7% AP50 on MBU benchmark with stable results.

Reliability-guided expert routing outperforms capacity-increased models.

Demonstrates robustness against synthetic shifts and complex backgrounds.

Abstract

Detecting small unmanned aerial vehicles from RGB-infrared remote-sensing pairs remains challenging due to tiny target scale, cluttered backgrounds, and spatial misalignment between heterogeneous sensors. Existing bimodal detectors often align or fuse features without assessing the reliability of local cross-sensor correspondence, allowing mismatch artifacts to propagate into the detection head. To address this issue, we propose LER-YOLO, a reliability-aware sparse mixture-of-experts framework for misaligned RGB-infrared UAV detection. LER-YOLO first introduces an Uncertainty-Aware Target Alignment module that resamples visible features toward the infrared reference and estimates a spatial reliability map. This reliability prior is then used by a Reliability-Guided Sparse MoE Fusion module to adaptively select k experts from RGB-dominant, infrared-dominant, and interactive fusion experts, enabling trustworthy cross-modal interaction while suppressing unreliable fusion. Experiments on the public MBU benchmark under a YOLOv5s-family protocol show that LER-YOLO achieves 89.7+/-0.2% AP50 over three independent seeds, with a best result of 89.9%. Extensive ablations, parameter-matched comparisons, synthetic-shift evaluations, and complexity analysis demonstrate that the gains mainly come from reliability-guided expert routing rather than increased model capacity.