Any3R: Feedforward 3D Reconstruction from Any Camera Type
Computer Science > Computer Vision and Pattern Recognition arXiv:2610.36150 (cs) [Submitted on 26 Oct 2026] Title: Any3R: Feedforward 3D Reconstruction from Any Camera Type Authors: Jingong Chen, Qingwen Zhang, Youming Deng, Chulwoo Pack, Olaf Wysocki, Boris Jutzi, Kwanghee Won, Kyle Gao.
计算机科学 > 计算机视觉与模式识别 arXiv:2610.36150 (cs) [提交于 2026 年 10 月 26 日] 标题：Any3R：适用于任何相机类型的前馈 3D 重建 作者：Jingong Chen, Qingwen Zhang, Youming Deng, Chulwoo Pack, Olaf Wysocki, Boris Jutzi, Kwanghee Won, Kyle Gao。
Abstract: Geometry foundation models recover camera poses and dense geometry from a set of images with a feed-forward network, but they expect pinhole images. That assumption is inside the backbone too: the positional encoding and the training loss both bind an image position to a direction in space, and both depend on the lens.
摘要：几何基础模型通过前馈网络从一组图像中恢复相机位姿和稠密几何结构，但它们通常预设为针孔相机图像。这种假设也存在于骨干网络中：位置编码和训练损失都将图像位置与空间方向绑定，且两者都依赖于镜头。
We condition both on the dense ray field, the unit direction through every pixel, which describes any central camera without a lens type, distortion coefficients or a bound on field of view. Angular coordinates give the rotary encoding the measured ray angle of each patch, evaluated per view, and solid-angle weighting scales the point loss by the area each pixel covers on the sphere, so neither the encoding nor the supervision depends on the lens.
我们将两者都以稠密射线场（即穿过每个像素的单位方向）为条件，这可以描述任何中心相机，而无需考虑镜头类型、畸变系数或视场限制。角坐标为旋转编码提供了每个图块的测量射线角度（按视图评估），而立体角加权则根据每个像素在球面上覆盖的面积来缩放点损失，因此编码和监督都不依赖于镜头。
Any3R is therefore a feed-forward model that takes any central camera, and is robust to mixing many camera types per scene, with no retraining per lens. On a benchmark of thirteen central cameras across three datasets, where every method reconstructs with its own predicted camera poses, it is the most consistently accurate method. Any3R enables scalable 3D reconstruction from heterogeneous crowdsourced and archival imagery collections sourced across diverse camera types.
因此，Any3R 是一个能够处理任何中心相机的前馈模型，对于同一场景中混合多种相机类型具有鲁棒性，且无需针对不同镜头进行重新训练。在涵盖三个数据集的 13 种中心相机的基准测试中，在所有方法均使用各自预测的相机位姿进行重建的情况下，它是最稳定准确的方法。Any3R 实现了从源自不同相机类型的异构众包和档案图像集合中进行可扩展的 3D 重建。