ROMEO: ROBUST METRIC VISUAL ODOMETRY

JunDa Cheng, zhipeng cai, Zhaoxing Zhang, Wei Yin, Matthias Müller, Michael Paulitsch, Xin Yang

Keywords: Visual odometry，monocular，SLAM

Abstract: Visual odometry (VO) aims to estimate camera poses from visual inputs --- the key for many applications such as VR/AR, robotics etc. This work focuses on monocular RGB VO where camera poses are directly estimated from a monocular RGB video without IMU or 3D sensors. Existing approaches lack robustness under this challenging scenario and fail to generalize to unseen data (especially outdoors); they also cannot recover metric-scale poses. Several methods have attempted to address these problems with priors from predicted depth. However, especially on unseen data, depth prediction noise can drastically degrade performance. We propose Robust Metric Visual Odometry (RoMeO), the first method that can leverage (noisy) depth priors to enable robust VO and recover metric scale poses. RoMeO incorporates both pre-trained monocular metric depth and multi-view stereo (MVS) models to recover metric-scale, simplify correspondence search, provide better initialization and regularize optimization. Effective strategies ensure the efficiency and the robustness to prior noise. RoMeO advances the state-of-the-art (SOTA) by a large margin across 6 diverse datasets covering both indoor and outdoor scenes. Compared to the current SOTA DPVO, RoMeO reduces the relative (align the trajectory scale with GT) and absolute trajectory errors on average by 55.2% and 77.8% respectively (Fig.1). The performance gain also transfers to the full SLAM pipeline (with global BA & loop closure). Code will be released upon acceptance.

Primary Area: applications to robotics, autonomy, planning

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 3174