[论文] Beyond Spatio-Temporal Priors: A Generalizable Approach for Dense Corr...

论文概要

研究领域: CV 作者: Luping Liu, Bingyi Kang, Yifan Wang, Dong Xu 发布时间: 2026-10-08 arXiv: 2610.12421

中文摘要

密集对应匹配历史上一直受限于简化的时空先验，如平滑运动和刚性几何。这些假设对经典任务有效，但在图像编辑和参考引导生成（IEG）中会失效——这些变换可以保持视觉身份但打破物理连续性。为在这类变换上建立保持身份的对应，我们引入 FreeMatching——一个将生成式和语义基础表征与来自经典数据集、跟踪视频和合成场景的异构监督相结合的通用框架。教师引导的迭代精炼进一步改善了 IEG 中的对应，无需密集对应标注。实验表明，单个 FreeMatching 模型在具有挑战性的 IEG 图像对上大幅提升了对应质量，同时在经典基准上保持竞争力。此外，我们展示了它作为评估身份保持的定量指标的效用，其分数与人类判断相关。代码：https://github.com/luping-liu/FreeMatching

原文摘要

Dense correspondence matching has historically been bounded by simplifying spatio-temporal priors, such as smooth motion and rigid geometry. While effective for classical tasks, these assumptions break down in image editing and reference-guided generation (IEG), where transformations can preserve visual identity while breaking physical continuity. To establish identity-preserving correspondence across such transformations, we introduce FreeMatching, a generalizable framework combining generative and semantic foundation representations with heterogeneous supervision from classical datasets, tracked videos, and synthetic scenes. Teacher-guided iterative refinement further improves correspondence in IEG without dense correspondence annotations. Experimentally, a single FreeMatching model subs...

#论文 #arXiv #CV #小凯