DCANet: Dense Context-Aware Network for Semantic Segmentation

TL;DR

DCANet introduces a Dense Context-Aware module that adaptively integrates local and global context information, enhancing feature representation for semantic segmentation.

Contribution

The paper proposes a novel DCA module and extended structures to better capture long-range dependencies, improving multi-scale feature aggregation in segmentation networks.

Findings

Achieves state-of-the-art performance on PASCAL VOC 2012

Demonstrates robustness on Cityscapes and ADE20K datasets

Improves multi-scale feature representation for semantic segmentation

Abstract

As the superiority of context information gradually manifests in advanced semantic segmentation, learning to capture the compact context relationship can help to understand the complex scenes. In contrast to some previous works utilizing the multi-scale context fusion, we propose a novel module, named Dense Context-Aware (DCA) module, to adaptively integrate local detail information with global dependencies. Driven by the contextual relationship, the DCA module can better achieve the aggregation of context information to generate more powerful features. Furthermore, we deliberately design two extended structures based on the DCA modules to further capture the long-range contextual dependency information. By combining the DCA modules in cascade or parallel, our networks use a progressive strategy to improve multi-scale feature representations for robust segmentation. We empirically demonstrate the promising performance of our approach (DCANet) with extensive experiments on three challenging datasets, including PASCAL VOC 2012, Cityscapes, and ADE20K.