Abstract
Weakly Supervised Semantic Segmentation (WSSS), which relies only on image-level labels, has attracted significant attention for its cost-effectiveness and scalability. Existing methods mainly enhance inter-class distinctions and employ data augmentation to mitigate semantic ambiguity and reduce spurious activations. However, they often neglect the complex contextual dependencies among image patches, resulting in incomplete local representations and limited segmentation accuracy. To address these issues, we propose the Context Patch Fusion with Class Token Enhancement (CPF-CTE) framework, which exploits contextual relations among patches to enrich feature representations and improve segmentation. At its core, the Contextual-Fusion Bidirectional Long Short-Term Memory (CF-BiLSTM) module captures spatial dependencies between patches and enables bidirectional information flow, yielding a more comprehensive understanding of spatial correlations. This strengthens feature learning and segmentation robustness. Moreover, we introduce learnable class tokens that dynamically encode and refine class-specific semantics, enhancing discriminative capability. By effectively integrating spatial and semantic cues, CPF-CTE produces richer and more accurate representations of image content. Extensive experiments on PASCAL VOC 2012 and MS COCO 2014 validate that CPF-CTE consistently surpasses prior WSSS methods.
Abstract (translated)
弱监督语义分割（WSSS）依赖于图像级别的标签，因其成本效益和可扩展性而引起了广泛关注。现有的方法主要通过增强类间区分度并采用数据增强技术来减轻语义模糊性和减少虚假激活。然而，它们往往忽视了图块之间的复杂上下文依赖关系，导致局部表示不完整且分割精度受限。为解决这些问题，我们提出了带有类别标记增强（CPF-CTE）的上下文图块融合框架，该框架利用图块之间的上下文关系来丰富特征表示并提高分割性能。在核心部分，上下文融合双向长短时记忆模块（Contextual-Fusion BiLSTM, CF-BiLSTM）捕获图块间的空间依赖性，并实现双向信息流动，从而对空间相关性提供更全面的理解。这增强了特征学习和分割的鲁棒性。此外，我们引入了可学习的类别标记，这些标记能够动态编码并细化特定类别的语义信息，提高了判别能力。通过有效整合空间和语义线索，CPF-CTE生成了更加丰富且准确的图像内容表示。 在PASCAL VOC 2012及MS COCO 2014数据集上的大量实验验证表明，CPF-CTE方法始终优于先前的WSSS方法。
URL
https://arxiv.org/abs/2601.14718
