论文/arXiv 2607.17171·3 个月前

VIDAR：通过几何基础模型的视觉-惯性稠密对齐与重建

VIDAR将SVO+IMU里程计与Depth Anything 3耦合，在EuRoC和TUM RGB-D验证单目度量稠密重建。

AI 整理

• 感知与导航
• 具身模型

原文链接：VIDAR: Visual-Inertial Dense Alignment and Reconstruction via a Geometric Foundation Model（arXiv 2607.17171 · 发布日期：2026-07-19）

VIDAR: Visual-Inertial Dense Alignment and Reconstruction via a Geometric Foundation Model

VIDAR将SVO+IMU里程计与Depth Anything 3耦合，在EuRoC和TUM RGB-D验证单目度量稠密重建。

核心要点

• 单目基础模型可提供稠密几何，但通常缺乏稳定的度量尺度。
• 视觉-惯性前端作为度量锚，提供相机位姿、尺度和一致世界坐标系。
• Depth Anything 3的局部几何预测跨时间对齐，并融合为全局重建。
• 在EuRoC上，位姿注入将尺度误差降至约1%，解耦混合mean F@0.10达0.676。

验证范围与局限：验证范围为EuRoC和TUM RGB-D上的单目度量稠密重建实验。

英文摘要 (Abstract)

Monocular foundation models provide dense geometry but usually lack a stable metric scale. This paper presents VIDAR, a visual-inertial dense reconstruction framework that couples SVO+IMU odometry with Depth Anything 3. VIDAR uses the visual-inertial front end as a metric anchor: it provides camera poses, scale, and a consistent world frame for aligning dense foundation-model predictions across time. The foundation model then contributes detailed local geometry that is fused into a global reconstruction. We study both pose-conditioned DA3 and a decoupled alignment strategy. On EuRoC, pose injection reduces scale error to about 1% and reaches 0.463 mean F@0.10; the decoupled hybrid improves this to 0.676 without ground-truth poses. Results on EuRoC and TUM RGB-D show that VIDAR is a practical route to metric dense monocular reconstruction.

作者：Diyari Mohammed Salih, Lingxiang Hu, Naïma Ait Oufroukh, Fabien Bonardi

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2607.17171：VIDAR: Visual-Inertial Dense Alignment and Reconstruction via a Geometric Foundation Model访问 2026-10-21