Saved in:
Bibliographic Details
Main Authors: Carreaud, Antoine, Naha, Elias, Chansel, Arthur, Lahellec, Nina, Skaloud, Jan, Gressin, Adrien
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.11310
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915930995425280
author Carreaud, Antoine
Naha, Elias
Chansel, Arthur
Lahellec, Nina
Skaloud, Jan
Gressin, Adrien
author_facet Carreaud, Antoine
Naha, Elias
Chansel, Arthur
Lahellec, Nina
Skaloud, Jan
Gressin, Adrien
contents Semantic ultra-high-resolution (UHR) image segmentation is essential in remote sensing applications such as aerial mapping and environmental monitoring. Transformer-based models remain challenging in this setting because memory grows quadratically with the number of tokens, limiting either spatial resolution or contextual scope. We introduce CASWiT (Context-Aware Stage-Wise Transformer), a dual-branch Swin-based architecture that injects low-resolution contextual information into fine-grained high-resolution features through lightweight stage-wise cross-attention. To strengthen cross-scale learning, we also propose a SimMIM-style pretraining strategy based on masked reconstruction of the high-resolution image. Extensive experiments on the large-scale FLAIR-HUB aerial dataset demonstrate the effectiveness of CASWiT. Under our RGB-only UHR protocol, CASWiT reaches 66.37% mIoU with a SegFormer decoder, improving over strong RGB baselines while also improving boundary quality. On the URUR benchmark, CASWiT reaches 49.2% mIoU under the official evaluation protocol, and it also transfers effectively to medical UHR segmentation benchmarks. Code and pretrained models are available at https://huggingface.co/collections/heig-vd-geo/caswit
format Preprint
id arxiv_https___arxiv_org_abs_2601_11310
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Context-Aware Semantic Segmentation via Stage-Wise Attention
Carreaud, Antoine
Naha, Elias
Chansel, Arthur
Lahellec, Nina
Skaloud, Jan
Gressin, Adrien
Computer Vision and Pattern Recognition
Semantic ultra-high-resolution (UHR) image segmentation is essential in remote sensing applications such as aerial mapping and environmental monitoring. Transformer-based models remain challenging in this setting because memory grows quadratically with the number of tokens, limiting either spatial resolution or contextual scope. We introduce CASWiT (Context-Aware Stage-Wise Transformer), a dual-branch Swin-based architecture that injects low-resolution contextual information into fine-grained high-resolution features through lightweight stage-wise cross-attention. To strengthen cross-scale learning, we also propose a SimMIM-style pretraining strategy based on masked reconstruction of the high-resolution image. Extensive experiments on the large-scale FLAIR-HUB aerial dataset demonstrate the effectiveness of CASWiT. Under our RGB-only UHR protocol, CASWiT reaches 66.37% mIoU with a SegFormer decoder, improving over strong RGB baselines while also improving boundary quality. On the URUR benchmark, CASWiT reaches 49.2% mIoU under the official evaluation protocol, and it also transfers effectively to medical UHR segmentation benchmarks. Code and pretrained models are available at https://huggingface.co/collections/heig-vd-geo/caswit
title Context-Aware Semantic Segmentation via Stage-Wise Attention
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2601.11310