HOPS：基于注意力感知过滤与亲和度引导增强的层次化开放词表部件分割
HOPS: Hierarchical Open-vocabulary Part Segmentation with Attention-Aware Filtering and Affinity-Guided Enhancement
将 CLIP 的语义对齐与 DINO 的结构感知通过双向注意力融合，并用注意力感知过滤抑制物体过分割、用亲和度引导增强缓解部件欠分割，这套两阶段流程可直接迁移到开放词表部件/属性分割与细粒度掩码生成任务。
原文依据 HOPS introduces a bidirectional semantic-structural attention fusion mechanism that integrates CLIP's semantic alignment with DINO's structural perception.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论开放词表部件分割（OVPS）旨在将物体分割为细粒度部件，同时泛化到未见类别。现有基于视觉语言模型的方法面临两个挑战：（1）物体过分割，由过于宽泛的语义激活引起；（2）部件欠分割，由细粒度感知能力薄弱导致。为解决这些问题，我们提出 HOPS，一个用于层次化开放词表部件分割的两阶段框架。HOPS 引入双向语义-结构注意力融合机制，将 CLIP 的语义对齐与 DINO 的结构感知相结合。在物体分割阶段，注意力感知过滤模块（AFM）通过语义-结构注意力细化跨模态相似度图，以抑制物体过分割。在部件分割阶段，亲和度引导增强模块（AEM）迭代传播部件响应以逐步扩展激活区域，有效缓解部件欠分割。在 Pascal-Part-116、ADE20K-Part-234 和 PartImageNet 上的实验表明，HOPS 取得了最优性能并具有更优的泛化能力。我们的代码可在 https://github.com/TJU-IDVLab/HOPS 获取。
英文原文摘要 1211 字符
Open-vocabulary part segmentation (OVPS) aims to segment objects into fine-grained parts while generalizing to unseen categories. Existing VLM-based methods face two challenges: (1) object over-segmentation, caused by overly broad semantic activations, and (2) part under-segmentation, resulting from weak fine-grained perception. To address these issues, we propose HOPS, a two-stage framework for hierarchical open-vocabulary part segmentation. HOPS introduces a bidirectional semantic-structural attention fusion mechanism that integrates CLIP's semantic alignment with DINO's structural perception. In the object segmentation stage, the Attention-Aware Filtering Module (AFM) refines cross-modal similarity maps via semantic-structural attention to suppress object over-segmentation. In the part segmentation stage, the Affinity-Guided Enhancement Module (AEM) iteratively propagates part responses to progressively expand activation regions, effectively mitigating part under-segmentation. Experiments on Pascal-Part-116, ADE20K-Part-234, and PartImageNet demonstrate that HOPS achieves state-of-the-art performance with superior generalization. Our code is available at https://github.com/TJU-IDVLab/HOPS.