So this is the VisualSFM version of the second Sleepy Cat model. I cleaned the point cloud, generated the mesh, and projected the textures in MeshLab. There are a few ups and downs with this compared to the MeshRoom version: Bad: The texture is dark and patchy Bad: There’s some lumpy reconstruction Bad: Far less detail Bad: Ears are lumpy Good: Fewer holes Good: Ears aren’t gone completely
I find it interesting that the number one issue I have with MeshRoom is that ears keep disappearing from everything. I suppose it’s a side effect of the depth map process, but I don’t know nearly enough about how it works. I just throw JPEGs into things and models come out!