To fully extract feature correlation information, we propose a Multi-scale Feature Attention Network (MFANet) for semantic segmentation. We design three key modules for our MFANet, including Channel Attention Module with two inputs (CAM2), Multi-scale Mixed Pooling (MMP) and Feature Attention Fusion (FAF) modules.
Fully extracting feature correlation between and within stages for semantic segmentation · Feiniu Yuan, Kang Li, +2 authors. Yaowen Zhu · Published in Digit.
Oct 22, 2024 · All the proposed modules use a way of attention mechanism to extract feature correlation between different stages and within the same stage.
Jul 1, 2022 · To fully extract correlation information, we design several special structures to propose a Multi-scale Feature Attention Network (MFANet).
Fully extracting feature correlation between and within stages for semantic segmentation ; Journal: Digital Signal Processing, 2022, p. 103578 ; Publisher: ...
Dec 11, 2024 · To solve the problem, this paper proposes a novel semantic segmentation method to improve the capacity of obtaining semantic information for the ...
Fully extracting feature correlation between and within stages for semantic segmentation. Article. May 2022; DIGIT SIGNAL PROCESS. Feiniu Yuan · Kang ...
People also ask
What is the semantic segmentation process?
What is semantic segmentation vs instance segmentation?
Dec 13, 2023 · We introduce a novel Region-Enhanced Feature Learning Network (REFL-Net) that leverages region correlations to enhance point feature learning.
This paper implements a fusion layer that allows to merge global features and local features in encoder network and introduces a stacked pooling block, ...
Jun 14, 2022 · Hierarchical feature extraction mainly realises the extraction of shallow low-level feature information and deep semantic feature information.