Converting a monocular video to dataset

[jmunshi1990]

Hi Team,

Thanks for the work. I am trying to use the xrdslam algorithms on a custom video from my work.

The video is a monocular video without any depth measurement. I would like to know how can I convert the video frames to a specific dataset type defined in xrdslam?

As far as I can see all the dataset type would need video frames, depth and pose. Can you tell me how to convert my video to a dataset which can be consumed by xrdslam for inference?

Thanks