MAF-Net: A Lightweight Morphology-Adaptive Feature Fusion Network for Semantic Segmentation.
- 1Guangling College, Yangzhou University, Yangzhou 225009, China.
- 2Kyushu Institute of Technology, Kitakyushu 8048550, Japan.
- 3School of Information and Artificial Intelligence, Yangzhou University, Yangzhou 225009, China.
Abstract
Semantic segmentation enables pixel-level scene understanding, and deploying it on resource-constrained platforms such as unmanned aerial vehicles and embedded sensing devices has become increasingly important. In complex scenes, however, objects with substantial scale variation and heterogeneous morphology often lead to missed small objects, discontinuities along slender structures, and incomplete boundaries. To address the challenges, this paper proposes a lightweight semantic segmentation network, named MAF-Net. The network adopts MobileNet as the backbone and enhances the representation of objects with different scales and morphological characteristics through multi-scale feature extraction and top-down feature fusion, while maintaining low model complexity. To further accommodate variations in object scale and structure, a Morphology-Adaptive Kernel Decoder (MAKD) is designed. The module introduces large-kernel, small-kernel, horizontal-strip, and vertical-strip convolution branches, and adaptively fuses the multi-branch features through a gating mechanism, thereby providing complementary contextual representations with different receptive-field scales and directional sensitivities. In addition, a Triple Feature Fusion Refinement Module (TFFRM) is constructed to further refine the initial segmentation results by integrating shallow spatial details, learned boundary-aware responses, and high-level semantic features, thereby alleviating problems such as discontinuities in slender structures and blurred object boundaries. Experimental results show that the proposed method achieves favorable overall segmentation performance on the COCO2017-derived ball-sports dataset, UDD6, and WHDLD, while maintaining a relatively small parameter count and high measured inference speed.
Related
- Lightweight SAR building segmentation via enhanced multi-scale context modeling and bidirectional guided feature fusion
- Cross-Directional Aggregation and Feature-Loss Optimization for Remote Sensing Semantic Segmentation
- Semantic-Guided Adaptive Gaussian Segmentation
- FasterGeo-YOLO: Multi-Scale Geometric Adaptive Feature Enhancement for Lightweight Road Damage Detection
- Agri-PIDNet-S: Lightweight semantic segmentation for field-structure and obstacle perception in seedling-stage cotton fields
BibTeX
@article{shi2026mafnet,
title = {{MAF-Net: A Lightweight Morphology-Adaptive Feature Fusion Network for Semantic Segmentation.}},
author = {Wenze Shi and Ping Liu and Takeshi Ikenaga and Daiki Nobayashi},
journal = {Sensors (Basel, Switzerland)},
year = {2026},
month = oct,
doi = {10.3390/s26206593},
url = {https://europepmc.org/article/MED/42944456},
note = {CC BY},
}
RIS
TY - JOUR
TI - MAF-Net: A Lightweight Morphology-Adaptive Feature Fusion Network for Semantic Segmentation.
AU - Wenze Shi
AU - Ping Liu
AU - Takeshi Ikenaga
AU - Daiki Nobayashi
PY - 2026
DA - 2026/10/17
JO - Sensors (Basel, Switzerland)
DO - 10.3390/s26206593
UR - https://europepmc.org/article/MED/42944456
AB - Semantic segmentation enables pixel-level scene understanding, and deploying it on resource-constrained platforms such as unmanned aerial vehicles and embedded sensing devices has become increasingly important. In complex scenes, however, objects with substantial scale variation and heterogeneous morphology often lead to missed small objects, discontinuities along slender structures, and incomplete boundaries. To address the challenges, this paper proposes a lightweight semantic segmentation network, named MAF-Net. The network adopts MobileNet as the backbone and enhances the representation of objects with different scales and morphological characteristics through multi-scale feature extraction and top-down feature fusion, while maintaining low model complexity. To further accommodate variations in object scale and structure, a Morphology-Adaptive Kernel Decoder (MAKD) is designed. The module introduces large-kernel, small-kernel, horizontal-strip, and vertical-strip convolution branches, and adaptively fuses the multi-branch features through a gating mechanism, thereby providing complementary contextual representations with different receptive-field scales and directional sensitivities. In addition, a Triple Feature Fusion Refinement Module (TFFRM) is constructed to further refine the initial segmentation results by integrating shallow spatial details, learned boundary-aware responses, and high-level semantic features, thereby alleviating problems such as discontinuities in slender structures and blurred object boundaries. Experimental results show that the proposed method achieves favorable overall segmentation performance on the COCO2017-derived ball-sports dataset, UDD6, and WHDLD, while maintaining a relatively small parameter count and high measured inference speed.
SN - 1424-8220
ER -