FrozenBurning / Text2Light

[SIGGRAPH Asia 2022] Text2Light: Zero-Shot Text-Driven HDR Panorama Generation

Home Page:https://frozenburning.github.io/projects/text2light/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

DataSet

xwg0508 opened this issue · comments

Hello, Thanks for your amazing work! About dataset , i'd like to ask about the size of pictures in the dataset, 1k,2k,4k or ? For example, Poly-Haven?

Thanks for your interest in our work. Please refer to #11 (comment) for details. The resolution of the envmap in the dataset varies from 1K to 4K as we collect from multiple sources. Still, we have the preprocessing script for you to handle data preprocessing before training.

Close due to inactivity. Feel free to reopen for further questions!