F2Net: A Frequency-Fused Network for Ultra-High Resolution Remote Sensing Segmentation 915 CCS: Computing methodologies Image segmentation
Hengzhi Chen, Liqian Feng, Wenhua Wu, Xiaogang Zhu, Shawn Leo, Kun Hu · 2025
Read F2Net: A Frequency-Fused Network for Ultra-High Resolution Remote Sensing Segmentation 915 CCS: Computing methodologies Image segmentation by Hengzhi Chen, Liqian Feng, Wenhua Wu, Xiaogang Zhu, Shawn Leo, Kun Hu (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-07847.txt.
Description
Hengzhi Chen and co-authors split satellite images by frequency to map land cover at extreme detail. Their 2025 paper tackles ultra-high-resolution scans, where shrinking loses fine lines and tiling loses the broad scene. Two-branch nets try both, but cost much and fight during training.
Their net sends fine detail down one branch at full size, and broad layout down another at small size with short- and long-range arms. A fusion block joins the reads, steered by two new losses: one aligns meaning across branches, one balances their training pull. On two known sets the method tops the field. City and land planners gain crisper maps.
Catalog record
| CREATOR | Hengzhi Chen, Liqian Feng, Wenhua Wu, Xiaogang Zhu, Shawn Leo, Kun Hu |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2506-07847