arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.03030cs.CV
收藏

基于前景-背景对比学习的夜间行人检测

Nighttime Pedestrian Detection Based on Fore-Background Contrast Learning

Guizhou University(贵州大学);North China Institute of Science and Technology(华北科技学院)

He Yao, Yongjun Zhang, Huachun Jian, Li Zhang, Ruzhong Cheng

AI总结:

本文提出前景-背景对比注意力机制 FBCA,通过将背景信息引入通道注意力并采用对比学习,提升单光谱夜间行人检测性能,在多个数据集上取得最优结果。

摘要:

在关于通道注意力机制的当代研究中,背景信息的重要性经常被忽视。本研究通过将背景信息纳入通道注意力机制,解决了低光照条件下单光谱夜间行人检测性能欠佳的问题。尽管已有大量研究致力于开发高效的通道注意力机制,但背景信息的相关性在很大程度上被忽略了。通过采用对比学习方法,我们重新审视了夜间行人检测中关于行人目标与背景信息的通道注意力,从而提出了前景-背景对比注意力(Fore-Background Contrast Attention,FBCA)。FBCA 具有两个主要特性:(1)通道描述符与全局空间特征信息形成远程依赖;(2)背景信息的整合增强了专注于低光照行人特征的通道与专注于背景信息的通道之间的区分度。因此,所获得的通道描述符表现出更高的语义层次和空间准确性。实验结果表明,FBCA 在单光谱夜间行人检测中显著优于现有方法,在 NightOwls 和 TJU-DHD-pedestrian 数据集上取得了最先进的结果。此外,该方法还为多光谱 LLVIP 数据集带来了性能提升。这些发现表明,将背景信息整合到通道注意力机制中可以有效缓解夜间场景中由光照因素引起的检测器性能下降。

英文摘要:

The significance of background information is frequently overlooked in contemporary research concerning channel attention mechanisms. This study addresses the issue of suboptimal single-spectral nighttime pedestrian detection performance under low-light conditions by incorporating background information into the channel attention mechanism. Despite numerous studies focusing on the development of efficient channel attention mechanisms, the relevance of background information has been largely disregarded. By adopting a contrast learning approach, we reexamine channel attention with regard to pedestrian objects and background information for nighttime pedestrian detection, resulting in the proposed Fore-Background Contrast Attention (FBCA). FBCA possesses two primary attributes: (1) channel descriptors form remote dependencies with global spatial feature information; (2) the integration of background information enhances the distinction between channels concentrating on low-light pedestrian features and those focusing on background information. Consequently, the acquired channel descriptors exhibit a higher semantic level and spatial accuracy. Experimental outcomes demonstrate that FBCA significantly outperforms existing methods in single-spectral nighttime pedestrian detection, achieving state-of-the-art results on the NightOwls and TJU-DHD-pedestrian datasets. Furthermore, this methodology also yields performance improvements for the multispectral LLVIP dataset. These findings indicate that integrating background information into the channel attention mechanism effectively mitigates detector performance degradation caused by illumination factors in nighttime scenarios.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑