SAPLNet: State-Aware Prototype Learning for Remote Sensing Segmentation

CGSNet: Category Prior-Guided Self-Supervised Semantic Segmentation for Remote Sensing Images

Multimodal fusion methods have shown great potential in remote sensing image analysis, but existing approaches rely heavily on massive amounts of annotated data. This is not only costly and time-consuming but also prone to subjective bias. To address this issue, we propose a category-prior-based self-supervised framework, CGSNet, which uses category prior maps extracted from multispectral images as supervisory signals for end-to-end training. An adaptive confidence-weighted pseudo-label generation mechanism is designed to alleviate noise and errors in prior maps by replacing binary labels with continuous confidence maps, enabling the learning of uncertain interclass features. In addition, a multispectral feature-guided refinement strategy utilizes color and texture information to calibrate class transition regions and enhance the discriminative power of pseudo-labels in complex scenes. A dynamic mask selection strategy further enhances the model’s robustness and generalization capabilities through progressive learning. Experiments demonstrate that CGSNet achieves state-of-the-art performance without the need for human annotation, achieving an Mean Intersection over Union (mIoU) score of 78.46% on the Gaofen image dataset (GID) (vegetation) dataset and 79.58% on the Zurich (vegetation) dataset—12.22% and 15.07% higher than existing methods, respectively—while exhibiting strong cross-dataset zero-shot generalization capabilities. The code will be available at https://github.com/NUAALISILab

MSPD-net: structural–appearance prototype decoupling for weakly supervised semantic segmentation

BASeg: Boundary-Aware Remote Sensing Segmentation with Structural Penalties

A Mahalanobis-Angle Boundary Loss (MABL) is proposed that explicitly enhances boundary and shape consistency and is introduced, built upon MABL, a boundary- aware remote sensing segmentation framework with Struc- tural Penalties.

Hierarchical Adaptive Feature Refinement Network for VHR Remote Sensing Image Segmentation

Semantic segmentation of very-high-resolution (VHR) remote sensing imagery increasingly benefits from strong pretrained hierarchical encoders, yet exploiting their multi-stage representations remains difficult. Nearby regions demand different balances between fine detail and semantic context, aggressive task-specific transformations perturb useful pretrained features, and conventional semantic supervision provides limited structural guidance. We present HAFR-Net, a progressive refinement framework that adaptively organizes and conservatively refines hierarchical representations instead of replacing them with a monolithic decoder transformation. Heterogeneity-Guided Stage-Adaptive Fusion (HG-SAF) predicts dense stage weights conditioned on local feature variation. A Frequency-Residual Adapter (FRA) then injects frequency information through a bounded, zero-initialized residual branch that keeps the fused representation as its reference. A Confusion-Aware Tri-Prior Decoder (CATP) finally regularizes the prediction with boundary, objectness, and training-derived class-relation cues. Under a matched Swin-B training and single-scale inference protocol, HAFR-Net attains 84.12%, 87.86%, 55.17%, and 67.70% mIoU on ISPRS Vaihingen, ISPRS Potsdam, LoveDA, and OpenEarthMap, improving the matched UPerNet baseline by 0.55, 0.95, 1.55, and 1.84 percentage points, respectively. Controlled analyses further show consistent spatial reweighting beyond content-only routing, improved boundary and thin-structure accuracy over matched spatial and spectral alternatives, and reduced confusion on pre-declared class pairs.

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

Pixel-level annotation of remote sensing imagery is costly, motivating weakly supervised semantic segmentation (WSSS) using only image-level labels. However, class activation maps (CAMs) often highlight only discriminative sub-regions and fail to separate adjacent land-cover regions, particularly in remote sensing scenes characterized by densely co-occurring land-cover classes and substantial variations in object scale. To address these limitations, we propose a three-stage framework that integrates complementary priors from Contrastive Language–Image Pre-training (CLIP), Self-Distillation with No Labels version 2 (DINOv2), and the Segment Anything Model (SAM). First, a lightweight CLIP adapter aligns vision–language priors with remote sensing imagery, while sigmoid-based multi-label decoupled distillation replaces class-competitive distillation with independent class-wise supervision, producing more complete CAMs. Second, DINOv2-guided feature clustering decomposes large merged regions before SAM prompt generation, while Spatial–Semantic Constraints are used to construct confidence-guided point-and-box prompts and reject excessively expanded or semantically inconsistent masks, thereby generating reliable pseudo-labels. Finally, a compact segmentation network is initialized with the weights learned in Stage 1 and retrained using the refined pseudo-labels generated in Stage 2, eliminating the need for foundation models during inference. Experiments on the Potsdam, LoveDA, and DeepGlobe datasets show that the proposed method achieves mean intersection over union (mIoU) scores of 53.16%, 52.66%, and 62.98%, respectively, outperforming state-of-the-art WSSS baselines by 6.55, 1.16, and 1.27 percentage points, respectively. These results demonstrate the effectiveness and generalizability of the proposed framework across diverse remote sensing scenarios under image-level supervision.

Text-Guided Dual Refinement for Domain Generalized Semantic Segmentation in Remote Sensing

Recently, domain generalized remote sensing semantic segmentation (DG-RSSS) methods leverage vision foundation models (VFMs) with a parameter-efficient fine-tuning (PEFT) strategy to achieve remarkable progress. Although VFMs offer robust representations under distribution shifts between different remote sensing scenes, they still have some limitations, including inaccurate segmentation between similar classes and boundary regions. To address these limitations, this work proposes a text-guided dual refinement (TGDR) approach for DG-RSSS, which contains a text-guided discriminative refinement (TDR) module and a text-guided mask features refinement (TMR) module. In particular, first, the proposed TDR module generates class-discriminated features by using class-related learnable tokens and interclass similarity to refine the original features from the frozen backbone, where the tokens are initialized with class texts. Second, the proposed TMR module injects class-specific semantics into the phase component of mask features by using class-related learnable tokens to enhance the correlation between the semantics of classes and scene contents in mask features for refining the prediction of class boundaries in scene contents. Extensive experiments demonstrate that the proposed TGDR approach achieves superior performance across multiple DG-RSSS benchmarks, e.g., achieving 65.6%, 54.0%, 59.0%, 45.2%, 48.8%, and 79.3% mIoU on the Potsdam–to–Vaihingen (P2V), Vaihingen–to–Potsdam (V2P), Rural–to–Urban (R2U), Urban–to–Rural (U2R), Aerial–to–Satellite II (A2S), and Satellite II–to–Aerial (S2A) benchmarks.