KleinPAT Dataset

Supplemental Material for Paper: KleinPAT: Optimal Mode Conflation For Time-Domain Precomputation Of Acoustic Transfer


Our KleinPAT algorithm is fast. Below you can find a few hundred sound models that we computed in about an hour on a single consumer GPU using KleinPAT. Each sound model contains surface vibrational modes and FFAT maps (acoustic transfer maps). You can play with them in our demo viewer.

The dataset is for research/education purposes. Original meshes are either from public repositories (e.g., Stanford scanning repo), or derived from Thingi10k. Models in Thingi10k are from thingiverse, where models can have various different licenses. Consult this page for more licensing information for each model. By downloading our model, you are agreeing to its terms.