计算机科学
卷积神经网络
分割
失败
变压器
人工智能
特征(语言学)
深度学习
编码(集合论)
块(置换群论)
背景(考古学)
卷积码
特征学习
模式识别(心理学)
深层神经网络
图像分割
源代码
代表(政治)
建筑
特征提取
人工神经网络
上下文模型
计算机视觉
空间语境意识
机器学习
网络体系结构
作者
Feixiang Du,Shengkun Wu,Xiang Wang,Aoxue Ding,Zhongliang Wang,Joel Chia Ming Than
出处
期刊:Science Progress
[SAGE Publishing]
日期:2026-01-01
卷期号:109 (1): 368504261419245-368504261419245
被引量:6
标识
DOI:10.1177/00368504261419245
摘要
In the past decade, convolutional neural networks (CNNs) and transformers have achieved wide application in semantic segmentation tasks. Although CNNs with transformer models greatly improve performance, the global context modeling remains inadequate. Recently, Mamba achieved great potential in vision tasks, showing its advantages in modeling long-range dependency. In this article, we propose a lightweight efficient CNN-Mamba network for semantic segmentation, dubbed as ECMNet. ECMNet combines CNN with Mamba skillfully in a capsule-based framework to address their complementary weaknesses. Specifically, we design an enhanced dual-attention block for a lightweight bottleneck. In order to improve the representation ability of the feature, we devise a multi-scale attention unit to integrate multi-scale feature aggregation, spatial aggregation, and channel aggregation. Moreover, a Mamba enhanced feature fusion module merges diverse level feature, significantly enhancing segmented accuracy. Extensive experiments on two representative datasets demonstrate that the proposed model excels in accuracy and efficiency balance, achieving 70.6% mIoU on Cityscapes and 73.6% mIoU on CamVid test datasets, with 0.87 M parameters and 8.27G FLOPs on a single RTX 3090 GPU platform. Source code will be available at https://github.com/feixiangdu/ECMNet.
科研通智能强力驱动
Strongly Powered by AbleSci AI