Nighttime traffic object detection via adaptively integrating event and frame domains

Intelligent perception is crucial in Intelligent Transportation Systems (ITS), with vision cameras as critical components. However, traditional RGB cameras exhibit a significant decline in performance when capturing nighttime traffic scenes, limiting their effectiveness in supporting ITS. In contras...

Full description

Saved in:
Bibliographic Details
Main Authors: Yu Jiang, Yuehang Wang, Minghao Zhao, Yongji Zhang, Hong Qi
Format: Article
Language:English
Published: KeAi Communications Co. Ltd. 2025-07-01
Series:Fundamental Research
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2667325823002376
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850066478572765184
author Yu Jiang
Yuehang Wang
Minghao Zhao
Yongji Zhang
Hong Qi
author_facet Yu Jiang
Yuehang Wang
Minghao Zhao
Yongji Zhang
Hong Qi
author_sort Yu Jiang
collection DOAJ
description Intelligent perception is crucial in Intelligent Transportation Systems (ITS), with vision cameras as critical components. However, traditional RGB cameras exhibit a significant decline in performance when capturing nighttime traffic scenes, limiting their effectiveness in supporting ITS. In contrast, event cameras possess a high dynamic range (140 dB vs. 60 dB for traditional cameras), enabling them to overcome frame degradation in low-light conditions. Recently, multimodal learning paradigms have made substantial progress in various vision tasks, such as image-text retrieval. Motivated by this progress, we propose an adaptive selection and fusion detection method that leverages both event and RGB frame domains to optimize nighttime traffic object detection jointly. To address the challenge of unbalanced multimodal data fusion, we design a learnable adaptive selection and fusion module. This module performs feature ranking and fusion in the channel dimension, allowing efficient multimodal fusion. Additionally, we construct a novel multi-level feature pyramid network based on multimodal attention fusion. This network extracts potential features to enhance robustness in detecting nighttime traffic objects. Furthermore, we curate a dataset for nighttime traffic scenarios comprising RGB frames and corresponding event streams. Through experiments, we demonstrate that our proposed method outperforms current state-of-the-art techniques in event-based, frame-based, and event and frame fusion methods. This highlights the effectiveness of integrating the event and frame domains in enhancing nighttime traffic object detection.
format Article
id doaj-art-79a2dc234a7f4dedbef5e9cdf10ea6ff
institution DOAJ
issn 2667-3258
language English
publishDate 2025-07-01
publisher KeAi Communications Co. Ltd.
record_format Article
series Fundamental Research
spelling doaj-art-79a2dc234a7f4dedbef5e9cdf10ea6ff2025-08-20T02:48:43ZengKeAi Communications Co. Ltd.Fundamental Research2667-32582025-07-01541633164410.1016/j.fmre.2023.08.004Nighttime traffic object detection via adaptively integrating event and frame domainsYu Jiang0Yuehang Wang1Minghao Zhao2Yongji Zhang3Hong Qi4The College of Computer Science and Technology, Jilin University, Changchun 130012, China; The Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun 130012, China; Corresponding author.The College of Computer Science and Technology, Jilin University, Changchun 130012, China; The Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun 130012, ChinaThe College of Earth Sciences, Jilin University, Changchun 130012, ChinaThe College of Computer Science and Technology, Jilin University, Changchun 130012, China; The Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun 130012, ChinaThe College of Computer Science and Technology, Jilin University, Changchun 130012, China; The Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun 130012, ChinaIntelligent perception is crucial in Intelligent Transportation Systems (ITS), with vision cameras as critical components. However, traditional RGB cameras exhibit a significant decline in performance when capturing nighttime traffic scenes, limiting their effectiveness in supporting ITS. In contrast, event cameras possess a high dynamic range (140 dB vs. 60 dB for traditional cameras), enabling them to overcome frame degradation in low-light conditions. Recently, multimodal learning paradigms have made substantial progress in various vision tasks, such as image-text retrieval. Motivated by this progress, we propose an adaptive selection and fusion detection method that leverages both event and RGB frame domains to optimize nighttime traffic object detection jointly. To address the challenge of unbalanced multimodal data fusion, we design a learnable adaptive selection and fusion module. This module performs feature ranking and fusion in the channel dimension, allowing efficient multimodal fusion. Additionally, we construct a novel multi-level feature pyramid network based on multimodal attention fusion. This network extracts potential features to enhance robustness in detecting nighttime traffic objects. Furthermore, we curate a dataset for nighttime traffic scenarios comprising RGB frames and corresponding event streams. Through experiments, we demonstrate that our proposed method outperforms current state-of-the-art techniques in event-based, frame-based, and event and frame fusion methods. This highlights the effectiveness of integrating the event and frame domains in enhancing nighttime traffic object detection.http://www.sciencedirect.com/science/article/pii/S2667325823002376Environmental perceptionIntelligent transportationNighttime object detectionEvent cameraMultimodal learning
spellingShingle Yu Jiang
Yuehang Wang
Minghao Zhao
Yongji Zhang
Hong Qi
Nighttime traffic object detection via adaptively integrating event and frame domains
Fundamental Research
Environmental perception
Intelligent transportation
Nighttime object detection
Event camera
Multimodal learning
title Nighttime traffic object detection via adaptively integrating event and frame domains
title_full Nighttime traffic object detection via adaptively integrating event and frame domains
title_fullStr Nighttime traffic object detection via adaptively integrating event and frame domains
title_full_unstemmed Nighttime traffic object detection via adaptively integrating event and frame domains
title_short Nighttime traffic object detection via adaptively integrating event and frame domains
title_sort nighttime traffic object detection via adaptively integrating event and frame domains
topic Environmental perception
Intelligent transportation
Nighttime object detection
Event camera
Multimodal learning
url http://www.sciencedirect.com/science/article/pii/S2667325823002376
work_keys_str_mv AT yujiang nighttimetrafficobjectdetectionviaadaptivelyintegratingeventandframedomains
AT yuehangwang nighttimetrafficobjectdetectionviaadaptivelyintegratingeventandframedomains
AT minghaozhao nighttimetrafficobjectdetectionviaadaptivelyintegratingeventandframedomains
AT yongjizhang nighttimetrafficobjectdetectionviaadaptivelyintegratingeventandframedomains
AT hongqi nighttimetrafficobjectdetectionviaadaptivelyintegratingeventandframedomains