Hi, beginner here, thanks for this amazing project, and sorry for bothering you.
So I'm not really understanding the usage page description about Mip-NeRF 360 dataset means.
"We support the original NeRF data format like [nerf-synthetic] and COLMAP dataset like [Mip-NeRF 360]"
In my understanding, it seems like first I need to use COLMAP to get my every frame's camera pose, then put them into nerf2mesh to calculate and get the results, right? And what about Mip-NeRF 360 dataset means? I know it is another extension of NeRF, so this means I can use the result I get from Mip-NeRF 360 and put it into nerf2mesh to get a more accurate mesh?
And in the custom dataset section you provided some commands according to the situation in filming, I see some command says for 360 captures, does it means the "360-degree" video or image? like fisheye, equirectangular.
Hi, beginner here, thanks for this amazing project, and sorry for bothering you.
So I'm not really understanding the usage page description about Mip-NeRF 360 dataset means. "We support the original NeRF data format like [nerf-synthetic] and COLMAP dataset like [Mip-NeRF 360]"
In my understanding, it seems like first I need to use COLMAP to get my every frame's camera pose, then put them into nerf2mesh to calculate and get the results, right? And what about Mip-NeRF 360 dataset means? I know it is another extension of NeRF, so this means I can use the result I get from Mip-NeRF 360 and put it into nerf2mesh to get a more accurate mesh?
And in the custom dataset section you provided some commands according to the situation in filming, I see some command says for 360 captures, does it means the "360-degree" video or image? like fisheye, equirectangular.