PoseFM: Relative Camera Pose Estimation Through Flow Matching
For autonomous navigation and augmented reality, this work introduces uncertainty-aware pose estimation, but the performance gains are incremental over existing methods.
PoseFM reformulates monocular frame-to-frame visual odometry as a generative task using Flow Matching, modeling camera motion as a distribution to enable uncertainty estimation. It achieves competitive absolute trajectory error on TartanAir, KITTI, and TUM-RGBD benchmarks, matching or outperforming existing methods on some trajectories.
Monocular visual odometry (VO) is a fundamental computer vision problem with applications in autonomous navigation, augmented reality and more. While deep learning-based methods have recently shown superior accuracy compared to traditional geometric pipelines, particularly in environments where handcrafted features struggle due to poor structure or lighting conditions, most rely on deterministic regression, which lacks the uncertainty awareness required for robust applications. We propose PoseFM, the first framework to reformulate monocular frame-to-frame VO as a generative task using Flow Matching (FM). By leveraging FM, we model camera motion as a distribution rather than a point estimate, learning to transform noise into realistic pose predictions via continuous-time ODEs. This approach provides a principled mechanism for uncertainty estimation and enables robust motion inference under challenging visual conditions. In our evaluations, PoseFM achieves strong performance on TartanAir, KITTI and TUM-RGBD benchmarks, achieving the lowest absolute trajectory error (ATE) on some of the trajectories and overall being competitive with the best frame-to-frame monocular VO methods. Code and model checkpoints will be made available at https://github.com/helsinki-sda-group/posefm.