efficient-sparse-voxel-octrees

Open-source implementation of "Efficient Sparse Voxel Octrees"

Proof-of-concept voxel rendering system presented in "Efficient Sparse Voxel Octrees" at I3D 2010. Utilizes the massively parallel computing power available in GPUs to implement real-time ray tracing of voxel data represented using a compact octree data structure.

See the sidebar for a download package with pre-compiled binaries, as well as a link to the paper. Feel free to post feedback on the discussion forum.

New in version 1.4

  • Switch to New BSD License (previously Apache License 2.0).
  • Upgrade to Visual Studio 2010 (previously 2008).
  • Support PNG textures through lodepng.
  • Fix a compiler error in the benchmark mode.
  • Fix a CUDA compilation issue with Visual Studio Express.
  • General bugfixes and improvements to framework.

System requirements

  • Microsoft Windows XP, Vista, or 7. 64-bit version is recommended to avoid running out of virtual address space when operating on large scenes.
  • At least 2 gigabytes of system memory.
  • NVIDIA CUDA-compatible GPU with compute capability 1.1 and at least 256 megabytes of DRAM. Quadro FX 5800 is recommended.
  • Microsoft Visual Studio 2010. Required even if you do not plan to build the source code, as the CUDA Toolkit depends on it.

Developers

  • Tero Karras
  • Samuli Laine

Contributors

  • Gregory J. Ward (conference room model)

Project Information

Labels:
Voxel Octree 3D GPU CUDA