跳到正文
返回论文列表
cs.CV提交于 已译

Accord:单相机与 IMU 的通用在线度量定位与稠密建图

Accord: Universal Online Metric Localization and Dense Mapping from one Camera and an IMU

Jaafar Mahmoud · Arthur Movsesyan · Cuong Huynh · Mikhail Iumanov · Sergey Zagoruyko · Sergey Kolyubin

中文摘要

仅凭单相机与 IMU 构建稠密度量地图仍具挑战:视觉惯性里程计(VIO)提供度量运动但几何稀疏,前馈深度模型预测稠密结构却无法保证度量尺度。Accord 能从任意带 IMU 的相机在线构建精确稠密度量地图。VIO 滤波器输出度量位姿,以此作为条件,深度基础模型对短关键帧窗口预测深度,因子图融合所得子图。关键在于多视图一致性(Accord),在 SLAM 中视为几何自洽来源。Accord 将每个关键帧窗口锚定到由滤波器位姿三角化得到的度量三维点,再利用共视图像间的双向稠密深度一致性估计并校正逐图像尺度误差,同时保留锚点建立的度量尺度。上下文关键帧进一步改善深度预测,所得子图在因子图中融合。在五个公开相机-IMU 配置及新 Phone-rig/GoPro 数据集上评估单一建图配置。无需按数据集/配置调参,且每张地图仅用 Accord 自身轨迹对齐,Accord 在每个平台上均取得 5 cm 平均 F-score 的最优/次优。Accord 与手机支架数据集已开源。

关键要点

  1. 01单相机+IMU 稠密度量建图:VIO 给度量位姿但几何稀疏,深度模型稠密但无度量尺度。
  2. 02用滤波器位姿三角化三维点锚定关键帧窗口,借共视图像双向深度一致性校正逐图尺度误差。
  3. 03上下文关键帧提升深度预测,子图在因子图中融合,保持度量尺度。
  4. 04五个公开相机-IMU 配置及新 Phone-rig/GoPro 数据集上,单一配置无需调参,5 cm 平均 F-score 均最优或次优。
  5. 05Accord 与手机支架数据集开源。

解读

尚无解读。

原始英文摘要

arXiv:2611.25279v1 Announce Type: new Abstract: Dense metric mapping from a single camera and IMU remains challenging: visual-inertial odometry (VIO) provides metric motion but sparse geometry, whereas feed-forward depth models predict dense structure without ensuring metric scale. We present Accord, which constructs an accurate online dense metric map from any camera with an IMU. A VIO filter publishes metric poses, a depth foundation model conditioned on them predicts depth for short keyframe windows, and a factor graph fuses the resulting submaps. Recent work has explored several approaches to this problem. What is essential is the multi-view agreement (Accord), which is considered in SLAM as a source of geometric self-consistency. Accord anchors each keyframe window to metric 3D points triangulated from the filter poses, then exploits bidirectional dense depth agreement between co-visible images to estimate and correct per-image scale errors while preserving the metric scale established by the anchors. Context keyframes further improve depth prediction, and the resulting submaps are fused in a factor graph. We evaluate a single mapping configuration across five public camera-IMU setups and a new Phone-rig/GoPro dataset. With nothing tuned per dataset/setup and every map aligned only by Accord's own trajectory, Accord achieves consistently best/second-best mean F-score at 5 cm on every platform. Accord is open-source along with the phone-rig dataset for the benefit of the research community

同方向论文 · cs.CV

查看全部 →