外观
论文arXiv 2609.217541 个月前
SFVO：置信度引导的解耦 Stereo-Flow 视觉里程计与双向 PnP
原文链接：SFVO: Decoupled Confidence-Guided Stereo-Flow Visual Odometry with Bidirectional PnP（arXiv 2609.21754 · 发布日期：2026-09-18）
SFVO: Decoupled Confidence-Guided Stereo-Flow Visual Odometry with Bidirectional PnP
SFVO利用预训练立体匹配与光流，将对应关系转为几何约束，在室内外数据集验证稳健准确位姿估计。
核心要点
- 单目VO存在尺度模糊，深度立体VO又受计算成本和建模复杂度限制。
- 框架直接利用预训练立体匹配和光流模型，估计立体与时间对应关系。
- 不从图像直接学习位姿，而将对应关系映射为几何约束并预测可信点。
- 室内外数据集实验显示其位姿估计稳健准确且泛化强，代码将会发布。
验证范围与局限：验证范围为室内外数据集上的位姿估计。
英文摘要 (Abstract)
Deep learning-based visual odometry (VO) has achieved significant progress, yet most existing methods focus on a monocular approach, which suffers from scale ambiguity. Stereo VO provides real metric by its nature, but remains less studied in deep learning VO due to its high computational cost and modeling complexity. Recent advances in stereo matching and optical flow estimation have made dense visual correspondence increasingly accurate and reliable, but their complementary geometric information has not been fully exploited for VO. In this paper, we present SFVO, a correspondence-driven stereo VO framework that directly builds upon pretrained stereo matching and optical flow models. SFVO exploits pretrained stereo matching and optical flow models to estimate stereo and temporal correspondences. Instead of learning pose directly from images, SFVO maps learned correspondences into geometric constraints and predicts which points are trustworthy. To improve the reliability of visual correspondence-based geometric constraints, we introduce decoupled confidence maps for rotation and translation. This design better aligns the characteristics of visual correspondence and 6-DoF transformations. Extensive experiments on outdoor and indoor datasets demonstrate that SFVO achieves robust and accurate pose estimation with strong generalization capability. The code will be released.
作者：Kai Zhang, Guoyang Zhao, Jun Ma
来源
本页事实对应的公开页面，以原文为准。