TNet: Terrace Convolutional Decoder Network for Remote Sensing Image Semantic Segmentation

TL;DR

TNet introduces a convolutional decoder that progressively fuses multi-resolution features for improved remote sensing image segmentation, emphasizing global-local integration with simple operations.

Contribution

It proposes a novel convolutional decoder architecture that effectively combines multi-resolution features without complex modules, enhancing segmentation performance.

Findings

Achieves 85.35% mIoU on ISPRS Vaihingen dataset.

Attains 87.05% mIoU on ISPRS Potsdam dataset.

Maintains high computational efficiency.

Abstract

In remote sensing, most segmentation networks adopt the UNet architecture, often incorporating modules such as Transformers or Mamba to enhance global-local feature interactions within decoder stages. However, these enhancements typically focus on intra-scale relationships and neglect the global contextual dependencies across multiple resolutions. To address this limitation, we introduce the Terrace Convolutional Decoder Network (TNet), a simple yet effective architecture that leverages only convolution and addition operations to progressively integrate low-resolution features (rich in global context) into higher-resolution features (rich in local details) across decoding stages. This progressive fusion enables the model to learn spatially-aware convolutional kernels that naturally blend global and local information in a stage-wise manner. We implement TNet with a ResNet-18 encoder (TNet-R) and evaluate it on three benchmark datasets. TNet-R achieves competitive performance with a mean Intersection-over-Union (mIoU) of 85.35\% on ISPRS Vaihingen, 87.05\% on ISPRS Potsdam, and 52.19\% on LoveDA, while maintaining high computational efficiency. Code is publicly available.