配额校准的细粒度对齐与上下文感知边缘分布用于文本行人检索
Quota-Calibrated Fine-Grained Alignment with Context-Aware Marginals for Text-based Person Retrieval
QC-Align 提出一种无需参数、即插即用的训练正则化器，通过上下文感知的边缘分布估计和配额校准的最优传输目标，显式约束文本词与视觉区域之间的多对多匹配质量，可被复用于任何需要细粒度跨模态对齐的视觉-语言任务（如跨模态检索、图文匹配），尤其在少样本和跨域场景下提升基线性能与可解释性。
原文依据 QC-Align is a parameter-free, plug-and-play training regularizer that requires no fine-grained annotations and incurs no inference overhead.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论基于文本的行人检索（TPR）的核心挑战在于建立文本词与视觉区域之间细粒度的多对多语义对齐。现有方法主要依赖逐点相似度或注意力机制，隐含地假设匹配是独立且平衡的。因此，在属性重叠和大量背景噪声的条件下，这些方法常常将匹配权重错误分配给非判别性区域或词，导致匹配结果模糊。为解决这一问题，我们提出了 QC-Align，一个由上下文感知边缘分布引导的配额校准细粒度对齐框架。具体而言，我们提出了一个上下文感知边缘估计器（CAME），动态地为每个词和视觉区域分配“匹配配额”，随后采用配额校准传输（QCT）目标来显式约束每个词和区域所能承载的匹配质量，从而在这些约束下联合优化文本与视觉之间的多对多对应关系。值得注意的是，QC-Align 是一个无参数、即插即用的训练正则化器，不需要细粒度标注，且不产生推理开销。在多个主流行人检索基准上的实验表明，QC-Align 持续提升基线模型性能，在少样本和跨域场景下增益更大且可解释性更好。
英文原文摘要 1422 字符
The core challenge in Text-based Person Retrieval (TPR) lies in establishing fine-grained, many-to-many semantic alignment between textual words and visual regions. Existing methods predominantly rely on pointwise similarity or attention mechanisms, implicitly assuming matches are independent and balanced. Consequently, under conditions of attribute overlap and substantial background noise, these methods often misallocate matching weights to non-discriminative regions or words, resulting in ambiguous matching outcomes. To address this, we propose QC-Align, a quota-calibrated fine-grained alignment framework guided by context-aware marginals. Specifically, we propose a Context-Aware Marginal Estimator (CAME) that dynamically assigns "matching quotas" to each word and visual region, and subsequently employs a Quota-Calibrated Transport (QCT) objective to explicitly constrain the matching quality each word and region can carry, thereby jointly optimizing the many-to-many correspondence between text and vision under these constraints. Notably, QC-Align is a parameter-free, plug-and-play training regularizer that requires no fine-grained annotations and incurs no inference overhead. Experiments on multiple mainstream person retrieval benchmarks demonstrate that QC-Align consistently improves baseline model performance, with greater gains and better interpretability in few-shot and cross-domain scenarios.