OpenVO：具备时序动态感知的开放世界视觉里程计
OpenVO: Open-World Visual Odometry with Temporal Dynamics Awareness
把观测频率作为显式条件编码进两帧位姿回归，并借助基础模型的3D几何先验，使单目视觉里程计在未知帧率与未标定相机下仍能输出真实尺度轨迹；做行车记录仪或任意变帧率视频的轨迹提取、3D重建时可直接复用这一设定。
原文依据 explicitly encodes temporal dynamics information within a two-frame pose regression framework
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论本文提出 OpenVO，一个在有限输入条件下具备时序感知能力的开放世界视觉里程计（VO）新框架。OpenVO 能够从观测频率多变、相机未标定的单目行车记录仪视频中有效估计真实世界尺度的自运动，从而支持从行车记录仪记录的稀有驾驶事件中稳健地构建轨迹数据集。现有 VO 方法在固定观测频率（如 10Hz 或 12Hz）下训练，完全忽略了时序动态信息。许多先前方法还要求相机已标定且内参已知。因此，当（1）部署于未见过的观测频率下，或（2）应用于未标定相机时，其性能会下降。这显著限制了它们对许多下游任务的泛化能力，例如从行车记录仪视频中提取轨迹。为应对这些挑战，OpenVO（1）在两帧位姿回归框架内显式编码时序动态信息，（2）利用源自基础模型的 3D 几何先验。我们在三个主要自动驾驶基准——KITTI、nuScenes 和 Argoverse 2——上验证了本方法，相较最优方法取得超过 20% 的性能提升。在观测频率变化的设置下，本方法显著更为稳健，在所有指标上误差降低 46%–92%。这些结果证明了 OpenVO 在真实世界 3D 重建及多样下游应用中的通用性。
英文原文摘要 1500 字符
We introduce OpenVO, a novel framework for Open-world Visual Odometry (VO) with temporal awareness under limited input conditions. OpenVO effectively estimates real-world-scale ego-motion from monocular dashcam footage with varying observation rates and uncalibrated cameras, enabling robust trajectory dataset construction from rare driving events recorded in dashcam.Existing VO methods are trained on fixed observation frequency (e.g., 10Hz or 12Hz), completely overlooking temporal dynamics information. Many prior methods also require calibrated cameras with known intrinsic parameters. Consequently, their performance degrades when (1) deployed under unseen observation frequencies or (2) applied to uncalibrated cameras. These significantly limit their generalizability to many downstream tasks, such as extracting trajectories from dashcam footage.To address these challenges, OpenVO (1) explicitly encodes temporal dynamics information within a two-frame pose regression framework and (2) leverages 3D geometric priors derived from foundation models. We validate our method on three major autonomous-driving benchmarks - KITTI, nuScenes, and Argoverse 2 - achieving more than 20% performance improvement over state-of-the-art approaches. Under varying observation rate settings, our method is significantly more robust, achieving 46%-92% lower errors across all metrics.These results demonstrate the versatility of OpenVO for real-world 3D reconstruction and diverse downstream applications.