SCTNet-NAS: efficient semantic segmentation via neural architecture search for cloud-edge collaborative perception

Abstract Efficient semantic segmentation on edge devices is critical for cloud-edge perception, yet achieving high accuracy under resource constraints remains challenging. To alleviate this problem, this paper presents SCTNet-NAS, a novel framework for efficient edge segmentation via cloud-edge co-o...

Full description

Saved in:
Bibliographic Details
Main Authors: Ruyu Liu, Lin Wang, Zhihao Yu, Haoyu Zhang, Xiufeng Liu, Bo Sun, Xv Huo, Jianhua Zhang
Format: Article
Language:English
Published: Springer 2025-06-01
Series:Complex & Intelligent Systems
Subjects:
Online Access:https://doi.org/10.1007/s40747-025-01996-5
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849235565478674432
author Ruyu Liu
Lin Wang
Zhihao Yu
Haoyu Zhang
Xiufeng Liu
Bo Sun
Xv Huo
Jianhua Zhang
author_facet Ruyu Liu
Lin Wang
Zhihao Yu
Haoyu Zhang
Xiufeng Liu
Bo Sun
Xv Huo
Jianhua Zhang
author_sort Ruyu Liu
collection DOAJ
description Abstract Efficient semantic segmentation on edge devices is critical for cloud-edge perception, yet achieving high accuracy under resource constraints remains challenging. To alleviate this problem, this paper presents SCTNet-NAS, a novel framework for efficient edge segmentation via cloud-edge co-optimization. SCTNet-NAS that unifies multi-objective neural architecture search (NAS), feature-based knowledge distillation from a cloud-based vision transformer (ViT) teacher, and specialized decoder design to simultaneously deliver high accuracy and real-time efficiency for semantic segmentation on resource-constrained edge devices. The method first constructs a weight-sharing supernet and applies an non-dominated sorting genetic algorithm (NSGA-II) to explore candidate encoders in a single forward pass, then transfers global context from a vision transformer teacher to each candidate via the VitGuidance feature-level distillation scheme. In addition, our meticulously designed SCTHead and AU_SCTHead decoder modules add adaptive channel re-calibration to enhance segmentation performance and refined boundary delineation. Evaluation demonstrates SCTNet-NAS achieves a significantly enhanced accuracy-efficiency trade-off versus state-of-the-art methods, enabling high-performance edge AI perception.
format Article
id doaj-art-8a6de3a3609149a7b1c1f34933681f46
institution Kabale University
issn 2199-4536
2198-6053
language English
publishDate 2025-06-01
publisher Springer
record_format Article
series Complex & Intelligent Systems
spelling doaj-art-8a6de3a3609149a7b1c1f34933681f462025-08-20T04:02:45ZengSpringerComplex & Intelligent Systems2199-45362198-60532025-06-0111811610.1007/s40747-025-01996-5SCTNet-NAS: efficient semantic segmentation via neural architecture search for cloud-edge collaborative perceptionRuyu Liu0Lin Wang1Zhihao Yu2Haoyu Zhang3Xiufeng Liu4Bo Sun5Xv Huo6Jianhua Zhang7School of Information Science and Technology, Hangzhou Normal UniversitySchool of Information Science and Technology, Hangzhou Normal UniversitySchool of Information Science and Technology, Hangzhou Normal UniversitySchool of Information Science and Technology, Hangzhou Normal UniversityDepartment of Technology, Management and Economics, Technical University of DenmarkHaixi Institutes, Chinese Academy of Sciences, Quanzhou Institute of Equipment ManufacturingSchool of Mathematics and Computer Science, Zhejiang A and F UniversitySchool of Computer Science and Engineering, Tianjin University of TechnologyAbstract Efficient semantic segmentation on edge devices is critical for cloud-edge perception, yet achieving high accuracy under resource constraints remains challenging. To alleviate this problem, this paper presents SCTNet-NAS, a novel framework for efficient edge segmentation via cloud-edge co-optimization. SCTNet-NAS that unifies multi-objective neural architecture search (NAS), feature-based knowledge distillation from a cloud-based vision transformer (ViT) teacher, and specialized decoder design to simultaneously deliver high accuracy and real-time efficiency for semantic segmentation on resource-constrained edge devices. The method first constructs a weight-sharing supernet and applies an non-dominated sorting genetic algorithm (NSGA-II) to explore candidate encoders in a single forward pass, then transfers global context from a vision transformer teacher to each candidate via the VitGuidance feature-level distillation scheme. In addition, our meticulously designed SCTHead and AU_SCTHead decoder modules add adaptive channel re-calibration to enhance segmentation performance and refined boundary delineation. Evaluation demonstrates SCTNet-NAS achieves a significantly enhanced accuracy-efficiency trade-off versus state-of-the-art methods, enabling high-performance edge AI perception.https://doi.org/10.1007/s40747-025-01996-5Semantic segmentationNeural architecture searchKnowledge distillationEdge computingCloud-edge collaboration
spellingShingle Ruyu Liu
Lin Wang
Zhihao Yu
Haoyu Zhang
Xiufeng Liu
Bo Sun
Xv Huo
Jianhua Zhang
SCTNet-NAS: efficient semantic segmentation via neural architecture search for cloud-edge collaborative perception
Complex & Intelligent Systems
Semantic segmentation
Neural architecture search
Knowledge distillation
Edge computing
Cloud-edge collaboration
title SCTNet-NAS: efficient semantic segmentation via neural architecture search for cloud-edge collaborative perception
title_full SCTNet-NAS: efficient semantic segmentation via neural architecture search for cloud-edge collaborative perception
title_fullStr SCTNet-NAS: efficient semantic segmentation via neural architecture search for cloud-edge collaborative perception
title_full_unstemmed SCTNet-NAS: efficient semantic segmentation via neural architecture search for cloud-edge collaborative perception
title_short SCTNet-NAS: efficient semantic segmentation via neural architecture search for cloud-edge collaborative perception
title_sort sctnet nas efficient semantic segmentation via neural architecture search for cloud edge collaborative perception
topic Semantic segmentation
Neural architecture search
Knowledge distillation
Edge computing
Cloud-edge collaboration
url https://doi.org/10.1007/s40747-025-01996-5
work_keys_str_mv AT ruyuliu sctnetnasefficientsemanticsegmentationvianeuralarchitecturesearchforcloudedgecollaborativeperception
AT linwang sctnetnasefficientsemanticsegmentationvianeuralarchitecturesearchforcloudedgecollaborativeperception
AT zhihaoyu sctnetnasefficientsemanticsegmentationvianeuralarchitecturesearchforcloudedgecollaborativeperception
AT haoyuzhang sctnetnasefficientsemanticsegmentationvianeuralarchitecturesearchforcloudedgecollaborativeperception
AT xiufengliu sctnetnasefficientsemanticsegmentationvianeuralarchitecturesearchforcloudedgecollaborativeperception
AT bosun sctnetnasefficientsemanticsegmentationvianeuralarchitecturesearchforcloudedgecollaborativeperception
AT xvhuo sctnetnasefficientsemanticsegmentationvianeuralarchitecturesearchforcloudedgecollaborativeperception
AT jianhuazhang sctnetnasefficientsemanticsegmentationvianeuralarchitecturesearchforcloudedgecollaborativeperception