发表机构
Beijing Institute of Technology; School of Computer Science, Wuhan University; Beihang University; INSAIT, Sofia University; BAAI AIR, Tsinghua University（北京理工大学; 武汉大学计算机科学学院; 北京航空航天大学; 索非亚大学INSAIT; 清华大学BAAI AIR）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文针对立体匹配和表面法线估计问题，提出统一框架GeoStereo，结合前馈立体匹配与基于扩散的法线估计分支，引入初始化策略与扭曲条件，实现有效交互，在多场景表现可靠，零样本设置下视差和法线估计精度高。
AI 中文摘要
立体匹配和表面法线估计是3D视觉中的基本任务。然而，现有的前馈立体方法在具有挑战性的区域仍难以产生可靠预测，主要是由于缺乏强大的几何先验。本文提出GeoStereo，一个统一的立体几何估计框架，利用强大的扩散先验联合预测视差和表面法线。具体而言，GeoStereo将前馈立体匹配管道与基于扩散的法线估计分支相结合。为实现两个任务间的有效交互，引入视差到法线的初始化策略，并为扩散过程构建到左视图的扭曲条件。这种耦合设计使扩散分支提供强大结构先验增强不适定区域的视差估计，前馈分支为准确的法线预测提供可靠几何指导。大量实验表明，GeoStereo在包括低光环境、高反射表面和透明物体等具有挑战性的场景中表现可靠。在零样本设置下，它在多个基准测试（如KITTI和NYUv2）上实现了视差估计的Rank-1，并在许多真实室内基准测试（如iBims-1和ScanNet）上提供了最佳的法线估计精度。
英文摘要
Stereo matching and surface normal estimation are fundamental tasks in 3D vision. However, existing feed-forward stereo methods still struggle to produce reliable predictions in challenging regions, mainly due to the lack of strong geometric priors. In this paper, we propose $\textbf{GeoStereo}$, a unified stereo geometry estimation framework that leverages powerful diffusion priors to jointly predict disparity and surface normals. Specifically, GeoStereo couples a feed-forward stereo matching pipeline with a diffusion-based normal estimation branch. To enable effective interaction between the two tasks, we introduce a disparity to normal initialization strategy and construct a warp to left-view condition for the diffusion process. This coupled design allows the diffusion branch to provide strong structural priors that enhance disparity estimation in ill-posed regions, while the feed-forward branch offers reliable geometric guidance for accurate normal prediction. Extensive experiments show that GeoStereo performs reliably in challenging scenarios, including low-light environments, highly reflective surfaces, and transparent objects. Under zero-shot settings, it achieves Rank-1 disparity estimation on multiple benchmarks, including KITTI and NYUv2, and delivers the best normal estimation accuracy on many real indoor benchmarks, such as iBims-1 and ScanNet. Project page: https://qz-wei.github.io/GeoStereo.github.io/
