超越语义搜索：面向组合图像检索的指代锚定
Beyond Semantic Search: Towards Referential Anchoring in Composed Image Retrieval
提出以边界框作为视觉锚点的细粒度组合图像检索任务，并设计上下文感知注意力调制器，可复用于需要实例级保真度的多模态检索与定位任务。
原文依据 we propose Object-Anchored Composed Image Retrieval (OACIR), a novel fine-grained retrieval task that mandates strict instance-level consistency.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论组合图像检索（CIR）通过支持结合参考图像与修改文本的灵活多模态查询，已展现出显著潜力。然而，CIR 本质上优先考虑语义匹配，难以在不同上下文中可靠地检索用户指定的实例。在实践中，强调具体实例保真度而非宽泛语义往往更为重要。在本工作中，我们提出对象锚定组合图像检索（OACIR），一种要求严格实例级一致性的新颖细粒度检索任务。为推进该任务的研究，我们构建了 OACIRR（真实图像上的 OACIR），这是首个大规模、多领域基准，包含超过 16 万个四元组和四个具有挑战性的候选库，并 enriched 了难负例实例干扰项。每个四元组通过一个边界框增强组合查询，该边界框在参考图像中视觉锚定对象，提供了一种精确且灵活的方式来确保实例保持。为解决 OACIR 任务，我们提出 AdaFocal，一个以上下文感知注意力调制器为特色的框架，该调制器自适应地增强指定实例区域内的注意力，动态平衡锚定实例与更广泛组合上下文之间的关注。大量实验表明，AdaFocal 显著优于现有的组合检索模型，特别是在保持实例级保真度方面，从而为这一挑战性任务建立了稳健的基线，同时为更灵活、实例感知的检索系统开辟了新方向。
英文原文摘要 1598 字符
Composed Image Retrieval (CIR) has demonstrated significant potential by enabling flexible, multimodal queries that combine a reference image and modification text. However, CIR inherently prioritizes semantic matching, struggling to reliably retrieve a user-specified instance across contexts. In practice, emphasizing concrete instance fidelity over broad semantics is often more consequential. In this work, we propose Object-Anchored Composed Image Retrieval (OACIR), a novel fine-grained retrieval task that mandates strict instance-level consistency. To advance research on this task, we construct OACIRR (OACIR on Real-world images), the first large-scale, multi-domain benchmark comprising over 160K quadruples and four challenging candidate galleries enriched with hard-negative instance distractors. Each quadruple augments the compositional query with a bounding box that visually anchors the object in the reference image, providing a precise and flexible way to ensure instance preservation. To address the OACIR task, we propose AdaFocal, a framework featuring a Context-Aware Attention Modulator that adaptively intensifies attention within the specified instance region, dynamically balancing focus between the anchored instance and the broader compositional context. Extensive experiments demonstrate that AdaFocal substantially outperforms existing compositional retrieval models, particularly in maintaining instance-level fidelity, thereby establishing a robust baseline for this challenging task while opening new directions for more flexible, instance-aware retrieval systems.