Question about table 1 and table 2

cwchenwang / NeRF-SR

NeRF-SR: High-Quality Neural Radiance Fields using Supersampling

https://cwchenwang.github.io/NeRF-SR

138 stars 13 forks source link

Question about table 1 and table 2 #3

Closed william2ai closed 2 years ago

william2ai commented 2 years ago

Hi, I'm very interested in your essay, I'd like to ask which specific dataset you'd used to hang out the result, like lego, chais, drums, etc. Or you just average the result. Thanks a lot!!

cwchenwang commented 2 years ago

We evaluated on two common datasets for NeRF: LLFF (contain scenes like fern) and Blender (contain scenes like lego). The results for both dataset are calculated by averaging the results for all the scenes in the dataset. Table 1 is for blender dataset and Table 2 is for LLFF dataset.

william2ai commented 2 years ago

ok, thanks a lot!