Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2504.16564 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909845627600896 |
|---|---|
| author | Wang, Zhongtao Cao, Xizhe Chen, Yisong Wang, Guoping |
| author_facet | Wang, Zhongtao Cao, Xizhe Chen, Yisong Wang, Guoping |
| contents | Semantic segmentation of remote sensing imagery demands precise spatial boundaries and robust intra-class consistency, challenging conventional hierarchical models. To address limitations arising from spatial domain feature fusion and insufficient receptive fields, this paper introduces SAIP-Net, a novel frequency-aware segmentation framework that leverages Spectral Adaptive Information Propagation. SAIP-Net employs adaptive frequency filtering and multi-scale receptive field enhancement to effectively suppress intra-class feature inconsistencies and sharpen boundary lines. Comprehensive experiments demonstrate significant performance improvements over state-of-the-art methods, highlighting the effectiveness of spectral-adaptive strategies combined with expanded receptive fields for remote sensing image segmentation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_16564 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SAIP-Net: Enhancing Remote Sensing Image Segmentation via Spectral Adaptive Information Propagation Wang, Zhongtao Cao, Xizhe Chen, Yisong Wang, Guoping Computer Vision and Pattern Recognition Graphics Semantic segmentation of remote sensing imagery demands precise spatial boundaries and robust intra-class consistency, challenging conventional hierarchical models. To address limitations arising from spatial domain feature fusion and insufficient receptive fields, this paper introduces SAIP-Net, a novel frequency-aware segmentation framework that leverages Spectral Adaptive Information Propagation. SAIP-Net employs adaptive frequency filtering and multi-scale receptive field enhancement to effectively suppress intra-class feature inconsistencies and sharpen boundary lines. Comprehensive experiments demonstrate significant performance improvements over state-of-the-art methods, highlighting the effectiveness of spectral-adaptive strategies combined with expanded receptive fields for remote sensing image segmentation. |
| title | SAIP-Net: Enhancing Remote Sensing Image Segmentation via Spectral Adaptive Information Propagation |
| topic | Computer Vision and Pattern Recognition Graphics |
| url | https://arxiv.org/abs/2504.16564 |