Reproduce “Neural Geometric Level of Detail: Real-time Rendering with Implicit 3D Shapes” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Towaki Takikawa, Joey Litalien, Kangxue Yin, Karsten Kreis, Charles Loop, Derek Nowrouzezahrai, Alec Jacobson, Morgan McGuire and 1 more
Neural Geometric LOD stores learned shape features in sparse voxel octrees and decodes them with a small multilayer perceptron. Its renderer selects detail levels and skips empty space. On Thingi32, it reports gIoU 99.4 and Chamfer-L1 0.0271 at LOD 5. In one 1920 × 1080 TurboSquid V Mech test, it takes 91 ms versus DeepSDF’s 10,843 ms, not a general real-time guarantee.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.