PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
April 3, 2026IEEE Transactions on Neural Networks and Learning Systems5 citations

Semantic Prompt and Graph-Convolution-Structure Distillation Framework for Semantic Segmentation of Remote Sensing Images

View Full Paper
WZWujie ZhouJXJin XieCXChunqiang Xu

Key Points

  • To develop an effective framework for high-resolution semantic segmentation of remote sensing images addressing challenges in modality and processing costs.
  • Proposed SPGSNet-S framework integrating multimodal feature enhancement and dual-path knowledge distillation.
  • Implemented auxiliary spatial feature extraction and RGB representation for feature alignment.
  • Introduced dual distillation scheme for capturing topological dependencies and generating visual prompts.
  • SPGSNet-S outperformed several state-of-the-art methods on Vaihingen and Potsdam datasets.
  • Achieved competitive performance with only 8.89 million parameters.
  • Maintained low computational cost with 2.29 G floating-point operations (FLOPs).

Abstract

High-resolution remote sensing semantic segmentation plays a critical role in land-use monitoring, urban planning, and disaster response. However, its deployment remains challenging owing to modality heterogeneity, fine-scale object structures, and the high computational cost of current deep learning models. To address these challenges, we propose a semantic prompt and graph-convolution-structure distillation framework (SPGSNet-S ^), a compact, yet effective architecture that integrates multimodal feature enhancement with dual-path knowledge distillation (KD). Specifically, we design two lightweight modules-auxiliary spatial feature extraction (ASFE) and red-green-blue (RGB) representation-to denoise and align noisy normalized digital surface model (nDSM) features with RGB imagery, enabling robust feature fusion. In addition, we introduce a dual distillation scheme comprising graph-convolution-based structure distillation, which captures and transfers spatial topological dependencies, and semantic prompt distillation (SPD), which dynamically generates and injects class-aware visual prompts without external text supervision. Experimental results on the Vaihingen and Potsdam datasets show that SPGSNet-S ^ outperforms several state-of-the-art methods, achieving competitive performance with only 8. 89 M parameters and 2. 29 G floating-point operations (FLOPs). The source code and experimental results are publicly available at https: //github. com/110-011/SPGSNet.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Zhou et al. (2026) studied this question.

synapsesocial.com/papers/69cf588f5a333a82146097a1https://doi.org/10.1109/tnnls.2026.3675381
Ask AI
Helpful
Bookmark
Share
View Full Paper