Question about the discrepancy between the paper's description and the code implementation.

[rockywind]

Hi, thank you for sharing the greak work. I was unable to find the corresponding code implementation for the approach mentioned in the quote below.

Our approach similarly partitions the scene into sub-NeRFs, making use of depth to improve partition efficiency and scaling over an area 200x larger than Block-NeRF’s Alamo Square Dataset. Both of these methods work only on static scenes

[hturki]

I didn't include it as part of the pre-release rewrite since we weren't able to release City-1M at the end of the day, but the code is pretty similar to that of Mega-NeRF (https://github.com/cmusatyalab/mega-nerf) except that we use depth instead of sampling along each ray. To be more precise, we used interpolated depth since our LiDAR depth maps are sparse.