F2Net：用于超高分辨率遥感分割的频率融合网络
F2Net: A Frequency-Fused Network for Ultra-High Resolution Remote Sensing Segmentation
针对超高分辨率图像分割，提出频率分解与多分支处理框架，通过跨频率对齐损失和平衡损失解决梯度冲突，可复用于其他需要同时保留细节和全局上下文的高分辨率视觉任务。
原文依据 We propose F2Net, a frequency-aware framework that decomposes UHR images into high- and low-frequency components for specialized processing.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论超高分辨率（UHR）遥感影像的语义分割对于环境监测和城市规划等应用至关重要，但面临计算和优化挑战。传统方法要么通过下采样丢失细节，要么通过图像块处理割裂全局上下文。虽然多分支网络解决了这一权衡，但存在计算效率低下和训练过程中梯度动态冲突的问题。我们提出F2Net，一个频率感知框架，将UHR图像分解为高频和低频分量进行专门处理。高频分支保留全分辨率结构细节，而低频分支通过两个子分支处理下采样输入，捕获短程和长程依赖关系。混合频率融合模块整合这些观测，并由两个新目标引导：跨频率对齐损失确保频率分量之间的语义一致性，跨频率平衡损失调节各分支的梯度幅度以稳定训练。在DeepGlobe和Inria Aerial基准上评估，F2Net分别以80.22和83.39的平均交并比达到最优性能。
英文原文摘要 1248 字符
Semantic segmentation of ultra-high-resolution (UHR) remote sensing imagery is critical for applications like environmental monitoring and urban planning but faces com- putational and optimization challenges. Conventional methods either lose fine details through downsampling or fragment global context via patch processing. While multi-branch networks ad- dress this trade-off, they suffer from computational inefficiency and conflicting gradient dynamics during training. We propose F2Net, a frequency-aware framework that decomposes UHR images into high- and low-frequency components for specialized processing. The high-frequency branch preserves full-resolution structural details, while the low-frequency branch processes downsampled inputs through dual sub-branches capturing short- and long-range dependencies. A Hybrid-Frequency Fusion mod- ule integrates these observations, guided by two novel objectives: Cross-Frequency Alignment Loss ensures semantic consistency between frequency components, and Cross-Frequency Balance Loss regulates gradient magnitudes across branches to stabilize training. Evaluated on DeepGlobe and Inria Aerial benchmarks, F2Net achieves state-of-the-art performance with mIoU of 80.22 and 83.39, respectively.