search

facebookresearch / localrf

An algorithm for reconstructing the radiance field of a large-scale scene from a single casually captured video.
MIT License
956 stars 62 forks source link

The division of training set and test set in Tanks&Temples dataset #49

Open Xiaohan-Z opened 4 months ago

Xiaohan-Z commented 4 months ago

Hello, thank you for your excellent work! But how do we divide the images in the Tanks&Temples dataset into training and test sets to reproduce the results in the paper? Anticipating for your earliest reply!

ameuleman commented 4 months ago

Hi, We preprocess Tanks and Temples scenes as follows: We get the full videos, retain one in every five frames as motion is slow and downscale them to full HD resolution (2048×1080 or 1920×1080), and keep the first 1000 images. Then, we use every tenth frame for testing (i.e. the first, the eleventh, ...). Please let me know if you need further clarification.