Skip to content

Releases: dsilvavinicius/m-plicits

Data and released models

Choose a tag to compare

@dsilvavinicius dsilvavinicius released this 16 Sep 21:10

Assets for reproducing the paper (unpack at the repository root, or use python tools/download_data.py):

  • m-plicits-data.zip (1.19 GB): clean inputs, 1% noisy variants, outlier clouds, the paper's checkpoints for all 36 shapes with their training configs, renderer and texture assets. SHA-256 99bd93725a8efcf9a2d60ed9c5e52b8b348369461d3ed33205a17b80d20310e5
  • m-plicits-paper-meshes.zip (0.32 GB, optional): the paper's reconstructions of the noisy inputs, for python reproduce.py paper-noise. SHA-256 ade561e88b1cbb53f80c9c6a2f28cd2ae73109435c3d7c58cb44d562ec95d933