频率感知亲和力用于弱监督语义分割
Frequency-Aware Affinity for Weakly Supervised Semantic Segmentation
利用频域分解（低频语义对齐、高频边界修正）生成更完整的亲和力矩阵，并通过最优传输直接生成高质量 CAM，可复用于其他弱监督定位任务（如弱监督目标检测、点监督分割）。
原文依据 we propose the Dual Frequency-Aware framework (DFA) to address this limitation
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论弱监督语义分割（WSSS）通常利用类激活图（CAM）提供像素级定位。然而，CAM 往往仅激活最具判别性的区域，导致 WSSS 性能欠佳。尽管现有的 CAM 细化方法利用亲和力中的成对关系来扩展激活区域，但这些源自视觉 Transformer（ViT）的亲和力表现出平滑特性，忽略了关键的高频关系，无法准确细化物体边界。在本工作中，我们提出双频率感知框架（DFA）来解决这一局限。具体而言，低频感知对齐（LFAA）生成低频感知亲和力，捕获显著的语义关系以增强 CAM 上物体内部的语义一致性，而高频感知修正（HFAR）模块产生高频感知亲和力，建模精确关系以保持 CAM 上物体边界结构。通过有效整合这两种互补的亲和力，我们基于最优传输理论设计了一种新颖的频率引导（FG）CAM 生成方法，显著省去了复杂的细化过程。大量实验表明，我们的 DFA 框架在 PASCAL VOC 和 MS COCO 基准上均取得了最优性能。代码将公开。
英文原文摘要 1374 字符
Weakly Supervised Semantic Segmentation (WSSS) typically utilizes Class Activation Maps (CAMs) to provide the pixel-wise localization. However, CAMs tend to activate only the most discriminative regions, leading to suboptimal WSSS performance. Although existing CAM refinement methods leverage pair-wise relations in affinity to expand the activation regions, these affinities derived from Vision Transformer (ViTs) exhibit a smoothing property, neglecting crucial high-frequency relations and failing to accurately refine object boundaries. In this work, we propose the Dual Frequency-Aware framework (DFA) to address this limitation. Specifically, the Low-Frequency-Aware Alignment (LFAA) generates low-frequency-aware affinity that captures salient semantic relations to enhance object interior semantic consistency on CAMs, while the High-Frequency-Aware Rectification (HFAR) module produces high-frequency-aware affinity that models precise relations to preserve object boundary structure on CAMs. By effectively integrating these two complementary affinities, we design a novel Frequency-Guided (FG) CAM Generation based on Optimal Transport theory, which significantly omits the complex refinement process. Extensive experiments demonstrate that our DFA framework achieves state-of-the-art performance on both PASCAL VOC and MS COCO benchmarks. Code will be released.