Stanford's Lucy

thingiverse

This exercise focused on reducing a large model with 28 million triangles down to just 460,000, resulting in a file size decrease from 1.4 GB to only 21 MB. The workflow involved sub-sampling the statue's points to 500K using CloudCompare and then performing surface reconstruction with Poisson in MeshLab (Octree Depth: 9). A netfabb repair was needed due to minor errors. For those interested, a script for this process is available in MeshLab. However, slicing at 25% scale might require 6 GB of RAM on the user's machine, making it impractical for models with less simplification. The original point-cloud file (.asc) and model are also included for those who want to try this method. The original model can be downloaded from Stanford CGL site.