从大型视觉模型中释放运动信息：面向步态识别的语义与运动学对偶框架
Unlocking Motion from Large Vision Models with a Semantic and Kinematic Duality for Gait Recognition
用视觉语言监督构造条件去相关损失（CDLoss），把步态嵌入中的衣物、视角等干扰因子显式解耦，这一思路可迁移到任何需要抑制 nuisance 因子的细粒度人体/行为表征学习；同时 GCaption 提供了带自然语言标注的步态数据集，可直接用于 context-aware 的人体分析。
原文依据 we introduce CDLoss, a Conditional Decorrelation Loss that explicitly disentangles the gait embeddings from nuisance factors using vision-language supervision
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论现有的基于集合的步态识别方法通过捕捉全局语义上下文取得了显著性能。然而，其顺序不变性使其无法对随时间展开的细粒度运动学模式进行建模。为统一全局表征与过程级表征，我们提出 GaitMax，一个同时捕捉语义上下文与运动学运动的框架。GaitMax 利用基于注意力的时空建模来动态表示细节化的部件级轨迹。这种细节化表征虽然更强大，但也会捕捉更多干扰因子（如衣物、视角），从而带来潜在的捷径问题。为缓解这一问题，我们引入 CDLoss，即条件去相关损失，它借助视觉语言监督显式地将步态嵌入与干扰因子解耦。该损失需要高质量的干扰描述。因此我们构建了 GCaption，这是一个为多个步态数据集提供自然语言标注的新资源，超越了简单的类别标签。GCaption 不仅支撑 CDLoss，还为未来上下文感知的步态分析奠定基础。GaitMax 的优越性通过在多个大规模步态基准上的大量实验得到验证。模型、代码与资源将在论文发表后公开。
英文原文摘要 1317 字符
Existing set-based gait recognition methods achieve remarkable performance by capturing global semantic context.However, their order-invariant nature prevents them from modeling the fine-grained kinematic patterns that unfold over time.To unify the global and process-level representations, we propose GaitMax, a framework that captures both semantic context and kinematic motion.GaitMax leverages attention-based spatiotemporal modeling to dynamically represent detailed part-level trajectories.While this detailed representation is more powerful, it also captures more nuisance factors (e.g., clothing, viewpoint), leading to potential shortcuts.To mitigate this, we introduce CDLoss, a Conditional Decorrelation Loss that explicitly disentangles the gait embeddings from nuisance factors using vision-language supervision.This loss requires high-quality nuisance descriptions. We therefore construct GCaption, a new resource that provides natural language annotations for multiple gait datasets, moving beyond simple categorical labels. GCaption not only enables CDLoss but also serves as a foundation for future context-aware gait analysis.The superiority of GaitMax is validated through extensive experiments on multiple large-scale gait benchmarks. Models, code, and resources will be released upon publication.