DSAC* Visual Re-Localization [Data]

Supplementary training data for visual camera re-localization, particularly rendered depth maps to be used in combination with the MSR 7Scenes dataset, and the Stanford 12Scenes dataset, as well as precomputed camera coordinate files for both aforementioned datasets. For more information, also see the code documentation: https://github.com/vislearn/dsacstar.

本数据集为视觉相机重定位任务的辅助训练数据，具体包含可与MSR 7场景数据集（MSR 7Scenes）、斯坦福12场景数据集（Stanford 12Scenes）配合使用的渲染深度图，以及针对上述两类数据集预先计算得到的相机坐标文件。如需更多信息，亦可参阅其代码文档：https://github.com/vislearn/dsacstar。