TUM-VIE

该数据集是一个高分辨率的以自我为中心的导航集，它将立体事件与视觉-惯性轨迹对齐。此外，该数据集提供了更高分辨率的传感器，以便进行精确的研究。其规模达到了亚厘米级的惯性姿态数据，任务包括深度估计和视觉-惯性同步定位与地图构建（SLAM）。

This dataset is a high-resolution egocentric navigation dataset that aligns stereo event data with visual-inertial trajectories. Additionally, it is equipped with higher-resolution sensors to enable precise research. It provides sub-centimeter-grade inertial attitude data, and the supported tasks include depth estimation and visual-inertial simultaneous localization and mapping (SLAM).