How to train custom dataset

facebookresearch / localrf

An algorithm for reconstructing the radiance field of a large-scale scene from a single casually captured video.

MIT License

976 stars 60 forks source link

How to train custom dataset #40

Closed t19cs008 closed 4 months ago

t19cs008 commented 1 year ago

Hi. Thank you for your work.

What should I prepare for training my custom dataset?

Your data in Google Drive contains images, sparse from colmap, and transforms.json from Instant-NGP script. Do I need to prepare sparse and transforms.json as well as images?

ameuleman commented 1 year ago

Hi, Instructions to prepare the dataset are available in the README. It only requires sorted image files in ${SCENE_DIR}/images. sparse/ and transforms.json are not needed when optimizing with default parameters.

xiaobiaodu commented 1 year ago

Hi @ameuleman

If I wanna train custom dataset, what fov I should set or I just set it as default?

ameuleman commented 1 year ago

The FoV depends on your focal length and sensor size. What camera are you using?