发表机构
Universitat Autònoma de Barcelona; Computer Vision Center（巴塞罗那自治大学; 计算机视觉中心）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出基于潜桥匹配的反照率估计架构，通过像素重建损失和阴影条件化解决物理一致性、推理成本及泛化问题，并在五个数据集上验证了其优于现有方法。
AI 中文摘要
内在图像分解（IID）的最新进展日益依赖于生成模型。然而，其进展仍受三个关键挑战的限制：（a）物理一致性不足，（b）推理时计算成本高，以及（c）泛化能力有限。在本工作中，我们表明潜桥匹配（LBM）能有效解决反照率估计中的这些局限。我们提出了一种新颖的基于LBM的架构，通过像素重建损失强制物理一致性，受益于LBM低成本推理的固有高效性，并通过引入阴影条件化来提升跨多样数据集的泛化能力。在此扩展版本中，我们进一步表明，使阴影估计器本身以预测的反照率为条件，能进一步提升重建保真度，并在五个真实与合成数据集上，将我们的最佳模型与最先进的IID方法进行基准比较。
英文摘要
Recent advances in Intrinsic Image Decomposition (IID) have increasingly relied on generative models. However, progress remains limited by three key challenges: (a) insufficient physical consistency, (b) high computational cost at inference time, and (c) limited generalization capabilities. In this work, we show that latent bridge matching (LBM) effectively addresses these limitations for albedo estimation. We introduce a novel LBM-based architecture that enforces physical consistency through a pixel reconstruction loss, benefits from the inherent efficiency of LBM low-cost inference, and improves generalization across diverse datasets by incorporating a shading conditioning. In this extended version, we additionally show that conditioning the shading estimator itself on the predicted albedo further improves reconstruction fidelity, and we benchmark our best model against stateof-the-art IID methods across five real and synthetic datasets.
CommentsAccpeted at the Color and Imaging Conference (CIC 2026), hosted by the Society for Imaging Science and Technology (IS&T)
