Hello,
Thank you very much for releasing this excellent work and the associated code. I am currently trying to strictly reproduce the sparse-view results reported in Table 1 / Table 3 on the Mip-NeRF360 dataset.
In Section 4, it is mentioned that for sparse-view novel view synthesis, three sparse-view subsets are selected per scene, and the detailed view selection policy is provided in the appendix. However, I could not find a precise description of how the 16-view subset is constructed.
I would like to kindly ask for clarification on the following points:
Under the 16-view setting on Mip-NeRF360, how are the views selected?
Uniformly spaced sampling?
Random sampling (with a fixed random seed)?
Or sampling based on camera pose distance?
Is the same view selection strategy (or same indices) used across all scenes?
Is there a public list of selected view indices or an official evaluation protocol available?
I would like to follow the experimental setup as strictly as possible for reproduction. Any clarification would be greatly appreciated.
Thank you again for your excellent work.
Hello,
Thank you very much for releasing this excellent work and the associated code. I am currently trying to strictly reproduce the sparse-view results reported in Table 1 / Table 3 on the Mip-NeRF360 dataset.
In Section 4, it is mentioned that for sparse-view novel view synthesis, three sparse-view subsets are selected per scene, and the detailed view selection policy is provided in the appendix. However, I could not find a precise description of how the 16-view subset is constructed.
I would like to kindly ask for clarification on the following points:
Under the 16-view setting on Mip-NeRF360, how are the views selected?
Uniformly spaced sampling?
Random sampling (with a fixed random seed)?
Or sampling based on camera pose distance?
Is the same view selection strategy (or same indices) used across all scenes?
Is there a public list of selected view indices or an official evaluation protocol available?
I would like to follow the experimental setup as strictly as possible for reproduction. Any clarification would be greatly appreciated.
Thank you again for your excellent work.