Discriminative Cross-Modal Attention Approach for RGB-D Semantic Segmentation

Scene understanding through semantic segmentation is a vital component for autonomous vehicles. Given the importance of safety in autonomous driving, existing methods are constantly striving to improve accuracy and reduce error. RGB-based semantic segmentation models typically underperform due to in...

Full description

Saved in:
Bibliographic Details
Main Authors: emad mousavian, Danial Qashqai, Shahriar B. Shokouhi
Format: Article
Language:English
Published: Ferdowsi University of Mashhad 2025-04-01
Series:Computer and Knowledge Engineering
Subjects:
Online Access:https://cke.um.ac.ir/article_46516_bbfb88302877289ce4d9c04dd311ac60.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849731683761258496
author emad mousavian
Danial Qashqai
Shahriar B. Shokouhi
author_facet emad mousavian
Danial Qashqai
Shahriar B. Shokouhi
author_sort emad mousavian
collection DOAJ
description Scene understanding through semantic segmentation is a vital component for autonomous vehicles. Given the importance of safety in autonomous driving, existing methods are constantly striving to improve accuracy and reduce error. RGB-based semantic segmentation models typically underperform due to information loss in challenging situations such as lighting variations and limitations in distinguishing occluded objects of similar appearance. Therefore, recent studies have developed RGB-D semantic segmentation methods by employing attention-based fusion modules. Existing fusion modules typically combine cross-modal features by focusing on each modality independently, which limits their ability to capture the complementary nature of modalities. To address this issue, we propose a simple yet effective module called the Discriminative Cross-modal Attention Fusion (DCMAF) module. Specifically, the proposed module performs cross-modal discrimination using element-wise subtraction in an attention-based approach. By integrating the DCMAF module with efficient channel- and spatial-wise attention modules, we introduce the Discriminative Cross-modal Network (DCMNet), a scale- and appearance-invariant model. Extensive experiments demonstrate significant improvements, particularly in predicting small and fine objects, achieving an mIoU of 77.39% on the CamVid dataset, outperforming state-of-the-art RGB-based methods, and a remarkable mIoU of 82.8% on the Cityscapes dataset. As the CamVid dataset lacks depth information, we employ the DPT monocular depth estimation model to generate depth images.
format Article
id doaj-art-05f0c35d30a5447385eda7e7bf20f434
institution DOAJ
issn 2538-5453
2717-4123
language English
publishDate 2025-04-01
publisher Ferdowsi University of Mashhad
record_format Article
series Computer and Knowledge Engineering
spelling doaj-art-05f0c35d30a5447385eda7e7bf20f4342025-08-20T03:08:28ZengFerdowsi University of MashhadComputer and Knowledge Engineering2538-54532717-41232025-04-0181435210.22067/cke.2025.88682.111746516Discriminative Cross-Modal Attention Approach for RGB-D Semantic Segmentationemad mousavian0Danial Qashqai1Shahriar B. Shokouhi2Department of Electrical Engineering, Iran University of Science and Technology, Tehran, Iran,Department of Electrical Engineering, Iran University of Science and Technology, Tehran, Iran,Department of Electrical Engineering, Iran University of Science and Technology, Tehran, IranScene understanding through semantic segmentation is a vital component for autonomous vehicles. Given the importance of safety in autonomous driving, existing methods are constantly striving to improve accuracy and reduce error. RGB-based semantic segmentation models typically underperform due to information loss in challenging situations such as lighting variations and limitations in distinguishing occluded objects of similar appearance. Therefore, recent studies have developed RGB-D semantic segmentation methods by employing attention-based fusion modules. Existing fusion modules typically combine cross-modal features by focusing on each modality independently, which limits their ability to capture the complementary nature of modalities. To address this issue, we propose a simple yet effective module called the Discriminative Cross-modal Attention Fusion (DCMAF) module. Specifically, the proposed module performs cross-modal discrimination using element-wise subtraction in an attention-based approach. By integrating the DCMAF module with efficient channel- and spatial-wise attention modules, we introduce the Discriminative Cross-modal Network (DCMNet), a scale- and appearance-invariant model. Extensive experiments demonstrate significant improvements, particularly in predicting small and fine objects, achieving an mIoU of 77.39% on the CamVid dataset, outperforming state-of-the-art RGB-based methods, and a remarkable mIoU of 82.8% on the Cityscapes dataset. As the CamVid dataset lacks depth information, we employ the DPT monocular depth estimation model to generate depth images.https://cke.um.ac.ir/article_46516_bbfb88302877289ce4d9c04dd311ac60.pdfattention mechanismautonomous drivingdeep learningrgb-d semantic segmentation
spellingShingle emad mousavian
Danial Qashqai
Shahriar B. Shokouhi
Discriminative Cross-Modal Attention Approach for RGB-D Semantic Segmentation
Computer and Knowledge Engineering
attention mechanism
autonomous driving
deep learning
rgb-d semantic segmentation
title Discriminative Cross-Modal Attention Approach for RGB-D Semantic Segmentation
title_full Discriminative Cross-Modal Attention Approach for RGB-D Semantic Segmentation
title_fullStr Discriminative Cross-Modal Attention Approach for RGB-D Semantic Segmentation
title_full_unstemmed Discriminative Cross-Modal Attention Approach for RGB-D Semantic Segmentation
title_short Discriminative Cross-Modal Attention Approach for RGB-D Semantic Segmentation
title_sort discriminative cross modal attention approach for rgb d semantic segmentation
topic attention mechanism
autonomous driving
deep learning
rgb-d semantic segmentation
url https://cke.um.ac.ir/article_46516_bbfb88302877289ce4d9c04dd311ac60.pdf
work_keys_str_mv AT emadmousavian discriminativecrossmodalattentionapproachforrgbdsemanticsegmentation
AT danialqashqai discriminativecrossmodalattentionapproachforrgbdsemanticsegmentation
AT shahriarbshokouhi discriminativecrossmodalattentionapproachforrgbdsemanticsegmentation