PedDet: Adaptive Spectral Optimization for Multimodal Pedestrian Detection

TL;DR

PedDet is a novel adaptive spectral optimization framework that enhances multimodal pedestrian detection by improving feature fusion and robustness to lighting variations, achieving state-of-the-art results in challenging conditions.

Contribution

Introduces PedDet, a framework with modules for adaptive spectral feature fusion and illumination robustness, advancing multispectral pedestrian detection performance.

Findings

Achieves 6.6% mAP improvement on LLVIP and MSDS datasets.

Demonstrates superior detection accuracy in low-light conditions.

Outperforms existing methods in complex scenarios.

Abstract

Pedestrian detection in intelligent transportation systems has made significant progress but faces two critical challenges: (1) insufficient fusion of complementary information between visible and infrared spectra, particularly in complex scenarios, and (2) sensitivity to illumination changes, such as low-light or overexposed conditions, leading to degraded performance. To address these issues, we propose PedDet, an adaptive spectral optimization complementarity framework specifically enhanced and optimized for multispectral pedestrian detection. PedDet introduces the Multi-scale Spectral Feature Perception Module (MSFPM) to adaptively fuse visible and infrared features, enhancing robustness and flexibility in feature extraction. Additionally, the Illumination Robustness Feature Decoupling Module (IRFDM) improves detection stability under varying lighting by decoupling pedestrian and background features. We further design a contrastive alignment to enhance intermodal feature discrimination. Experiments on LLVIP and MSDS datasets demonstrate that PedDet achieves state-of-the-art performance, improving the mAP by 6.6% with superior detection accuracy even in low-light conditions, marking a significant step forward for road safety. Code will be available at https://github.com/AIGeeksGroup/PedDet.