|On this page|
The Volume Resample operation resamples the voxels of a volume to a new resolution.
This is similar to scaling an image in COPs, so similar filtering operations are provided.
The volume bounding box will be unchanged by this operation. This means that unless the new resolution is an integer multiple of the old, the resulting voxels may no longer be square.
This node currently only works with standard Houdini volumes. It does not work with VDBs.
The volume primitives to be resampled.
Specifies the type of filter to use when scaling. The Box filter is the fastest. The Gaussian is typically the best choice for most uses. Catmull is used when you need something a bit sharper than Gaussian.
The default scale is set to properly capture the information in the original volume. Increasing this results in a blurring effect and decreasing may cause it to start to ignore incoming voxel information.
Specify Exact Resolution
The final voxel resolution may either be computed from the size of the box or set as a ratio to the incoming resolution.
This parameter controls which axis is divided into Uniform Sampling Divs.
Uniform Sampling Divs
The number of voxels to divide the Uniform Sampling axis into. The other axes will be divided into the number of cells that fit for this voxel size.
When using non-uniform voxel cells, each dimension’s resolution can be specified here.
The uniform size of the voxels, when specifying the voxel size directly. The given sized box will be filled by voxels of this size.
The ratio of the original resolution to use. 2.0 will result in twice as many voxels in each dimension, for a total of 8 times the voxels.
Autodetect 2d Volumes
2D volumes are kept in their original plane and not resampled along their thin axis. Sometimes, however, one may have a 3d volume that has collapsed to 2d, so want it to still resample in the third dimension.
This example shows how to use the Volume Resample SOP to increase and decrease the resolution of a volume. It also shows how the different sampling options affect the quality of the resulting volume.
The following examples include this node.