Background-Aware Cross-Attention Multiscale Fusion for Multispectral Object Detection

Limited by the imaging capabilities of sensors, research based on single modality is difficult to cope with faults and dynamic perturbations in detection. Effective multispectral object detection, which can achieve better detection accuracy by fusing visual information from different modalities, has...

Full description

Saved in:
Bibliographic Details
Main Authors: Runze Guo, Xiaojun Guo, Xiaoyong Sun, Peida Zhou, Bei Sun, Shaojing Su
Format: Article
Language:English
Published: MDPI AG 2024-10-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/16/21/4034
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850062863113125888
author Runze Guo
Xiaojun Guo
Xiaoyong Sun
Peida Zhou
Bei Sun
Shaojing Su
author_facet Runze Guo
Xiaojun Guo
Xiaoyong Sun
Peida Zhou
Bei Sun
Shaojing Su
author_sort Runze Guo
collection DOAJ
description Limited by the imaging capabilities of sensors, research based on single modality is difficult to cope with faults and dynamic perturbations in detection. Effective multispectral object detection, which can achieve better detection accuracy by fusing visual information from different modalities, has attracted widespread attention. However, most of the existing methods adopt simple fusion mechanisms, which fail to utilize the complementary information between modalities while lacking the guidance of a priori knowledge. To address the above issues, we propose a novel background-aware cross-attention multiscale fusion network (BA-CAMF Net) to achieve adaptive fusion in visible and infrared images. First, a background-aware module is designed to calculate the light and contrast to guide the fusion. Then, a cross-attention multiscale fusion module is put forward to enhance inter-modality complement features and intra-modality intrinsic features. Finally, multiscale feature maps from different modalities are fused according to background-aware weights. Experimental results on LLVIP, FLIR, and VEDAI indicate that the proposed BA-CAMF Net achieves higher detection accuracy than the current State-of-the-Art multispectral detectors.
format Article
id doaj-art-5ccc75389bc64ff0833d591e67866145
institution DOAJ
issn 2072-4292
language English
publishDate 2024-10-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-5ccc75389bc64ff0833d591e678661452025-08-20T02:49:49ZengMDPI AGRemote Sensing2072-42922024-10-011621403410.3390/rs16214034Background-Aware Cross-Attention Multiscale Fusion for Multispectral Object DetectionRunze Guo0Xiaojun Guo1Xiaoyong Sun2Peida Zhou3Bei Sun4Shaojing Su5College of Intelligence Science and Technology, National University of Defense Technology, Changsha 410073, ChinaCollege of Intelligence Science and Technology, National University of Defense Technology, Changsha 410073, ChinaCollege of Intelligence Science and Technology, National University of Defense Technology, Changsha 410073, ChinaCollege of Intelligence Science and Technology, National University of Defense Technology, Changsha 410073, ChinaCollege of Intelligence Science and Technology, National University of Defense Technology, Changsha 410073, ChinaCollege of Intelligence Science and Technology, National University of Defense Technology, Changsha 410073, ChinaLimited by the imaging capabilities of sensors, research based on single modality is difficult to cope with faults and dynamic perturbations in detection. Effective multispectral object detection, which can achieve better detection accuracy by fusing visual information from different modalities, has attracted widespread attention. However, most of the existing methods adopt simple fusion mechanisms, which fail to utilize the complementary information between modalities while lacking the guidance of a priori knowledge. To address the above issues, we propose a novel background-aware cross-attention multiscale fusion network (BA-CAMF Net) to achieve adaptive fusion in visible and infrared images. First, a background-aware module is designed to calculate the light and contrast to guide the fusion. Then, a cross-attention multiscale fusion module is put forward to enhance inter-modality complement features and intra-modality intrinsic features. Finally, multiscale feature maps from different modalities are fused according to background-aware weights. Experimental results on LLVIP, FLIR, and VEDAI indicate that the proposed BA-CAMF Net achieves higher detection accuracy than the current State-of-the-Art multispectral detectors.https://www.mdpi.com/2072-4292/16/21/4034multispectral object detectioncomplementary informationpriori knowledgebackground awarecross attentionmultiscale fusion
spellingShingle Runze Guo
Xiaojun Guo
Xiaoyong Sun
Peida Zhou
Bei Sun
Shaojing Su
Background-Aware Cross-Attention Multiscale Fusion for Multispectral Object Detection
Remote Sensing
multispectral object detection
complementary information
priori knowledge
background aware
cross attention
multiscale fusion
title Background-Aware Cross-Attention Multiscale Fusion for Multispectral Object Detection
title_full Background-Aware Cross-Attention Multiscale Fusion for Multispectral Object Detection
title_fullStr Background-Aware Cross-Attention Multiscale Fusion for Multispectral Object Detection
title_full_unstemmed Background-Aware Cross-Attention Multiscale Fusion for Multispectral Object Detection
title_short Background-Aware Cross-Attention Multiscale Fusion for Multispectral Object Detection
title_sort background aware cross attention multiscale fusion for multispectral object detection
topic multispectral object detection
complementary information
priori knowledge
background aware
cross attention
multiscale fusion
url https://www.mdpi.com/2072-4292/16/21/4034
work_keys_str_mv AT runzeguo backgroundawarecrossattentionmultiscalefusionformultispectralobjectdetection
AT xiaojunguo backgroundawarecrossattentionmultiscalefusionformultispectralobjectdetection
AT xiaoyongsun backgroundawarecrossattentionmultiscalefusionformultispectralobjectdetection
AT peidazhou backgroundawarecrossattentionmultiscalefusionformultispectralobjectdetection
AT beisun backgroundawarecrossattentionmultiscalefusionformultispectralobjectdetection
AT shaojingsu backgroundawarecrossattentionmultiscalefusionformultispectralobjectdetection