search

zju3dv / ENeRF

SIGGRAPH Asia 2022: Code for "Efficient Neural Radiance Fields for Interactive Free-viewpoint Video"
https://zju3dv.github.io/enerf
Other
413 stars 28 forks source link

Where the code confuses me #26

Closed JACKYLUO1991 closed 1 year ago

JACKYLUO1991 commented 1 year ago

Why lib/datasets/zjumocap/enerf.py $input_views_num need to plus 1?

haotongl commented 1 year ago

Hi, thanks for your attention.

During training, ENeRF randomly sample $input_views_num images for nearby views to render the target view. $input_views_num plus 1 is the number of nearby views. We make it slightly bigger than $input_views_num.