FFVO: A Feedforward Pose Decoder for Long-Horizon Visual Odometry

2026-09-12

AI Summary

为解决长视频中姿态估计的计算成本、上下文模糊性和时间不稳定性问题，提出FFVO方法，通过紧凑表示、层次解码器及轨迹监督实现高效稳定姿态估计。

Abstract

Stable and reliable 4D spatial understanding is fundamental for autonomous driving systems. While feedforward reconstruction networks can estimate camera motion and 3D structure in one pass, pose estimation over long videos remains challenged by computational cost, long-context ambiguity, and temporal instability. To address these challenges, we propose Feedforward Visual Odometry (FFVO), a pose-specialized adaptation of joint reconstruction architectures for efficient and temporally stable camera-pose estimation. FFVO uses (i) a compact camera-token representation for computationally efficient temporal aggregation, (ii) a hierarchical local-to-global temporal decoder that mitigates geometric ambiguity by separating short-range motion aggregation from sequence-level integration, and (iii) intermediate trajectory supervision that promotes temporal stability. Extensive evaluation on the Waymo Open Dataset (WOD), KITTI, and a large-scale proprietary benchmark demonstrates that our method performs favorably against existing feedforward approaches, and greatly reduces jitter and drift. These results support FFVO as an effective feedforward camera-pose decoder in long-horizon visual odometry settings.

Problem

• visual odometry
• pose estimation
• temporal stability

Innovation

• Feedforward Visual Odometry
• camera-token representation
• hierarchical local-to-global temporal decoder
• intermediate trajectory supervision