You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello, I'm encountering an error. During the "merge pointcloud.ply" step following the single train of a certain chunk, I get the error throw std::runtime_error("Found Nans!"). This error usually doesn't occur, but it suddenly appeared with a new dataset. Do you have any insights on this?
The text was updated successfully, but these errors were encountered:
When using a new dataset, have you paid attention to the degree of the SH coefficients (SHS)? During the hierarchy-building process, the PLY file will be parsed. If the degree of the SH coefficients is inconsistent, the memory addresses of subsequent variables like scale will be misaligned. This will directly lead to incorrect values for weight and covariance, when merging gaussians.
Hello, I'm encountering an error. During the "merge pointcloud.ply" step following the single train of a certain chunk, I get the error throw std::runtime_error("Found Nans!"). This error usually doesn't occur, but it suddenly appeared with a new dataset. Do you have any insights on this?
The text was updated successfully, but these errors were encountered: