Skip to content
Open access

Dual-Level Prototype Alignment via Cross-Attention for Few-Shot Remote Sensing Semantic Segmentation

2026 · Jordanian Journal of Computers and Information Technology · 0 citations

TL;DR

DLPANet is proposed, a novel dual-level prototype alignment network centered on Prototype-Guided Spatial Attention, enabling simultaneous modeling of scene context and fine-grained details and demonstrates that the decoupled dual cross-attention mechanism provides superior prototype-query alignment compared to prior global-local strategies.

Abstract

Analyzing remote sensing imagery relies heavily on accurate pixel-level classification to interpret complex geographical scenes. However, training robust deep learning models demands densely annotated datasets, which are expensive and time-consuming to create. Few-shot semantic segmentation (FSS) provides a practical alternative by adapting to new categories from limited labeled examples. Despite success in natural images, direct application to remote sensing faces unique challenges, such as massive scale variations, strong visual similarities among densely packed objects, and severe background interference. Existing FSS methods typically rely on masked average pooling for prototype extraction, which discards critical spatial geometries and fine-grained details. While some approaches attempt multi-scale or global-local modeling, they lack effective, decoupled alignment between support prototypes and query features, limiting their ability to simultaneously capture broad scene context and precise object boundaries in cluttered aerial views. To address these limitations, we propose DLPANet, a novel dual-level prototype alignment network centered on Prototype-Guided Spatial Attention (PGSA). Our framework constructs global and semantic prototypes from hierarchical backbone features, enabling simultaneous modeling of scene context and fine-grained details. Dual cross-attention modules then align these prototypes with decoupled global and semantic query feature maps, facilitating dynamic and precise feature enhancement that better distinguishes adjacent objects and suppresses background ambiguity. A learnable decoder integrates the enriched features, optimized via Dice loss to handle extreme class imbalance. Evaluated on the iSAID-5 benchmark, DLPANet achieves 36.12% and 41.05% mIoU in 1-shot and 5-shot settings, respectively, consistently outperforming state-of-the-art methods. These results demonstrate that our decoupled dual cross-attention mechanism provides superior prototype-query alignment compared to prior global-local strategies. https://github.com/mansoorfateh83-lab/dlpanet

Read PDF

Similar papers

Open access Aug 2026

Weakly Supervised Remote Sensing Segmentation via Decoupled Cross-Modal Distillation and Semantic-Guided Refinement

A three-stage framework that integrates complementary priors from Contrastive Language–Image Pre-training, Self-Distillation with No Labels version 2 (DINOv2), and the Segment Anything Model (SAM) is proposed, demonstrating the effectiveness and generalizability of the proposed framework across diverse remote sensing s...

Jing Li, Yu-Lin Cao, Xian-Tao Jiang et al. · 0 citations
#artificial intelligence Preprint Sep 2026

SatOV: Restoring Spatial Priors for Training-Free Open-Vocabulary Segmentation in Remote Sensing Imagery

SatOV is proposed, a training-free framework for open-vocabulary remote sensing segmentation that restores spatial priors at two stages of the representation pipeline that consistently improves training-free OVS and achieves competitive quantitative and qualitative results against state-of-the-art methods.

Chang-Hao Zhao, Ling-Lin Zeng, Hai Liu · 0 citations
Open access Aug 2026

Shared multi-branch framework and enhanced feature fusion for few-shot semantic segmentation

Experimental results demonstrate that the method significantly narrows the performance gap in data-constrained scenarios, achieving a mean Intersection over Union increase of 4.6% in one-shot and 5.5% in five-shot settings over the baseline.

Jin-Ming Guo, Li-Hsuan Chen, Yi-Chong Zeng et al. · 0 citations
Open access 2026

NRMLC: A Noise-Robust Multilabel Remote Sensing Scene Classification Framework Based on Labelwise Attention and Transformer-Based Classification

Multilabel remote sensing scene classification (ML-RSSC) aims to assign multiple semantic labels to complex high-resolution scenes. Although exhaustive expert annotation enables accurate supervision, its high cost limits large-scale applications. Low-cost labels from crowdsourcing, existing geographic products, or weak...

Xiao-Man Qi, Xing-Rui Zeng, Xiao-Tong Qi et al. · 0 citations
Open access Sep 2026

HDSMNet: Height-Guided Sparse Cross-Modal Fusion for High-Resolution Remote Sensing Semantic Segmentation

HDSMNet is proposed, a dual-branch multimodal semantic segmentation network designed for optical–nDSM data that enhances discriminative dense feature representations in high-resolution remote sensing images through interaction with a compact set of geometry-guided anchors.

Han-Xun Gu, Jiang-Jie Hu, Li Wang et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.