Context-Aware Semantic Segmentation via Stage-Wise Attention

• URL: http://arxiv.org/abs/2601.11310v1
• Date: Fri, 16 Jan 2026 14:06:46 GMT
• Title: Context-Aware Semantic Segmentation via Stage-Wise Attention
• Authors: Antoine Carreaud, Elias Naha, Arthur Chansel, Nina Lahellec, Jan Skaloud, Adrien Gressin,
• Abstract summary: We introduce CASWiT (Context-Aware Stage-Wise Transformer), a dual-branch, Swin-based architecture that injects global cues into fine-grained UHR features.<n>A cross-scale fusion module, combining cross-attention and gated feature injection, enriches high-resolution tokens with context.<n>Experiments on the large-scale IGN FLAIR-HUB aerial dataset demonstrate the effectiveness of CASWiT.
• Score: 0.10995326465245929
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Semantic ultra high resolution image (UHR) segmentation is essential in remote sensing applications such as aerial mapping and environmental monitoring. Transformer-based models struggle in this setting because memory grows quadratically with token count, constraining either the contextual scope or the spatial resolution. We introduce CASWiT (Context-Aware Stage-Wise Transformer), a dual-branch, Swin-based architecture that injects global cues into fine-grained UHR features. A context encoder processes a downsampled neighborhood to capture long-range dependencies, while a high resolution encoder extracts detailed features from UHR patches. A cross-scale fusion module, combining cross-attention and gated feature injection, enriches high-resolution tokens with context. Beyond architecture, we propose a SimMIM-style pretraining. We mask 75% of the high-resolution image tokens and the low-resolution center region that spatially corresponds to the UHR patch, then train the shared dual-encoder with small decoder to reconstruct the UHR initial image. Extensive experiments on the large-scale IGN FLAIR-HUB aerial dataset demonstrate the effectiveness of CASWiT. Our method achieves 65.83% mIoU, outperforming RGB baselines by 1.78 points. On URUR, CASWiT achieves 49.1% mIoU, surpassing the current SoTA by +0.9% under the official evaluation protocol. All codes are provided on: https://huggingface.co/collections/heig-vd-geo/caswit.

Related papers

• SegEarth-R1: Geospatial Pixel Reasoning via Large Language Model [61.97017867656831]
  We introduce a new task, ie, geospatial pixel reasoning, which allows implicit querying and reasoning and generates the mask of the target region.<n>We construct and release the first large-scale benchmark dataset called EarthReason, which comprises 5,434 manually annotated image masks with over 30,000 implicit question-answer pairs.<n>SegEarth-R1 achieves state-of-the-art performance on both reasoning and referring segmentation tasks, significantly outperforming traditional and LLM-based segmentation methods.
  arXiv  Detail & Related papers  (2025-04-13T16:36:47Z)
• AerOSeg: Harnessing SAM for Open-Vocabulary Segmentation in Remote Sensing Images [21.294581646546124]
  AerOSeg is a novel Open-Vocabulary (OVS) approach for remote sensing data.<n>We compute robust image-text correlation features using rotated versions of the input image and domain-specific prompts.<n>Inspired by the success of the Segment Anything Model (SAM) in diverse domains, we leverage SAM features to guide the spatial refinement of correlation features.<n>We enhance the refined correlation features using a multi-scale attention-aware composition to produce the final segmentation map.
  arXiv  Detail & Related papers  (2025-04-12T13:06:46Z)
• LKASeg:Remote-Sensing Image Semantic Segmentation with Large Kernel Attention and Full-Scale Skip Connections [27.473573286685063]
  We propose a remote-sensing image semantic segmentation network named LKASeg. LKASeg combines Large Kernel Attention(LSKA) and Full-Scale Skip Connections(FSC) On the ISPRS Vaihingen dataset, the mF1 and mIoU scores achieved 90.33% and 82.77%.
  arXiv  Detail & Related papers  (2024-10-14T12:25:48Z)
• Low-Resolution Self-Attention for Semantic Segmentation [93.30597515880079]
  We introduce the Low-Resolution Self-Attention (LRSA) mechanism to capture global context at a significantly reduced computational cost.<n>Our approach involves computing self-attention in a fixed low-resolution space regardless of the input image's resolution.<n>We demonstrate the effectiveness of our LRSA approach by building the LRFormer, a vision transformer with an encoder-decoder structure.
  arXiv  Detail & Related papers  (2023-10-08T06:10:09Z)
• OPE-SR: Orthogonal Position Encoding for Designing a Parameter-free Upsampling Module in Arbitrary-scale Image Super-Resolution [11.74426147465809]
  Implicit neural representation (INR) is a popular approach for arbitrary-scale image super-resolution. We propose an OPE-Upscale module to replace the INR-based upsampling module for arbitrary-scale image super-resolution.
  arXiv  Detail & Related papers  (2023-03-02T09:26:14Z)
• Attention Attention Everywhere: Monocular Depth Prediction with Skip Attention [6.491470878214977]
  Monocular Depth Estimation (MDE) aims to predict pixel-wise depth given a single RGB image. Inspired by the demonstrated benefits of attention in a multitude of computer vision problems, we propose an attention-based fusion of encoder and decoder features.
  arXiv  Detail & Related papers  (2022-10-17T13:14:47Z)
• Learning to Aggregate Multi-Scale Context for Instance Segmentation in Remote Sensing Images [28.560068780733342]
  A novel context aggregation network (CATNet) is proposed to improve the feature extraction process. The proposed model exploits three lightweight plug-and-play modules, namely dense feature pyramid network (DenseFPN), spatial context pyramid ( SCP), and hierarchical region of interest extractor (HRoIE)
  arXiv  Detail & Related papers  (2021-11-22T08:55:25Z)
• Small Lesion Segmentation in Brain MRIs with Subpixel Embedding [105.1223735549524]
  We present a method to segment MRI scans of the human brain into ischemic stroke lesion and normal tissues. We propose a neural network architecture in the form of a standard encoder-decoder where predictions are guided by a spatial expansion embedding network.
  arXiv  Detail & Related papers  (2021-09-18T00:21:17Z)
• UltraSR: Spatial Encoding is a Missing Key for Implicit Image Function-based Arbitrary-Scale Super-Resolution [74.82282301089994]
  In this work, we propose UltraSR, a simple yet effective new network design based on implicit image functions. We show that spatial encoding is indeed a missing key towards the next-stage high-accuracy implicit image function. Our UltraSR sets new state-of-the-art performance on the DIV2K benchmark under all super-resolution scales.
  arXiv  Detail & Related papers  (2021-03-23T17:36:42Z)
• Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers [149.78470371525754]
  We treat semantic segmentation as a sequence-to-sequence prediction task. Specifically, we deploy a pure transformer to encode an image as a sequence of patches. With the global context modeled in every layer of the transformer, this encoder can be combined with a simple decoder to provide a powerful segmentation model, termed SEgmentation TRansformer (SETR) SETR achieves new state of the art on ADE20K (50.28% mIoU), Pascal Context (55.83% mIoU) and competitive results on Cityscapes.
  arXiv  Detail & Related papers  (2020-12-31T18:55:57Z)
• Bi-directional Cross-Modality Feature Propagation with Separation-and-Aggregation Gate for RGB-D Semantic Segmentation [59.94819184452694]
  Depth information has proven to be a useful cue in the semantic segmentation of RGBD images for providing a geometric counterpart to the RGB representation. Most existing works simply assume that depth measurements are accurate and well-aligned with the RGB pixels and models the problem as a cross-modal feature fusion. In this paper, we propose a unified and efficient Crossmodality Guided to not only effectively recalibrate RGB feature responses, but also to distill accurate depth information via multiple stages and aggregate the two recalibrated representations alternatively.
  arXiv  Detail & Related papers  (2020-07-17T18:35:24Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.