search

JiahuiLei / MoSca

MIT License
99 stars 0 forks source link

Images split on Nvidia Dynamic datasets #4

Open kcheng1021 opened 1 month ago

kcheng1021 commented 1 month ago

Hi, thanks for your great work.

I notice that you follow roboust dynamic radiance field to split the nvidia dynamic datasets. But I found the data downloaded from roboust dynamic radiance field(https://github.com/facebookresearch/robust-dynrf) only contains 11 images for each scene. Is this true?

It seems different from the split from DynIbar (https://github.com/google/dynibar).

JiahuiLei commented 3 weeks ago

Thanks for your question and sorry for our late reply, there are quite a lot of traveling happened recently.

yes, we follow the RoDyNeRF since it's the closest baseline that does COLMAP-free dynamic rendering. And indeed they just use 11 frames to train and many following up works inherited such a setting, so we also use 11 frames for reporting the results.

hope this would be helpful.