Welcome to fVDB!

fVDB, inspired by function notation to resemble $f(VDB)$, is a data structure for encoding and operating on sparse voxel hierarchies of features in PyTorch. A sparse voxel hierarchy is a coarse-to-fine hierarchy of sparse voxel grids such that every fine voxel is contained within some coarse voxel. fvdb supports storing PyTorch tensors at the corners and centers of voxels in a hierarchy and enables a number of differentiable operations on these tensors (e.g. trilinear interpolation, splatting, ray tracing).

Please refer to the tutorials for examples of how to install and use fVDB to build your own sparse voxel pipelines.

Tutorials

• Basic Concepts
• Building Sparse Grids
• Basic GridBatch Operations
• A Simple Convolutional U-Net
• Sparse Grid I/O
• Mutable Grids
• Volume Rendering

API References

• GridBatch
• JaggedTensor

• fvdb.nn
  • VDBTensor
  • MaxPool
  • AvgPool
  • UpsamplingNearest
  • FillFromGrid
  • SparseConv3d
  • GroupNorm
  • Linear
  • ReLU
  • LeakyReLU
  • SELU
  • SiLU
  • Tanh
  • Sigmoid
  • Dropout
• fvdb.utils
  • FVDBExtension()

---

Indices and tables

• Index

• Module Index

• Search Page