These two models are identical and uses the same material without any textures. The only difference is the vertex normals.
•The head on the Left has the typical shading of having its normals set to each polygon face. This can create a very blocky look that may not well portray the object the voxels are supposed to represent. A common option is to make the scene or material totally shadeless and don’t react to lighting.
•The head on the Right instead has it’s normals derived from its original pre-voxelized model and its normals set uniformly for each voxel cube. I think it looks cool, allows the voxels the represent the object well and allows it to be shaded nicely by light sources.
•Another common voxel shading mode is to have each voxel uniformly average the light it would receive on all of its exposed faces. It would make the shading make more sense when you have lone floating voxels or very thin parts of the model etc. Maybe it can be achived here by overlapping many semi-transparent instances of cubes.
4 comments