F-Hash: Feature-Based Hash Design for Time-Varying Volume Visualization via Multi-Resolution Tesseract Encoding

F-Hash is a novel feature-based multi-resolution Tesseract encoding architecture to greatly enhance the convergence speed compared with existing input encoding methods for modeling time-varying volumetric data. The proposed design incorporates multi-level collision-free hash functions that map dynamic 4D multi-resolution embedding grids without bucket waste, achieving high encoding capacity with compact encoding parameters. Our encoding method is agnostic to time-varying feature detection methods, making it a unified encoding solution for feature tracking and evolution visualization.

Github Page, ArXiv, Publishers' Version

Demo video can be found here.

1. Packages

pip install vtk

2. Install Tinycudann

Get the Tinycudann source from here. Default Tinycudann supporting half-precision. To support Float 32 go into include/tiny-cuda-nn/common.h and change

#define TCNN_HALF_PRECISION (!(TCNN_MIN_GPU_ARCH == 61 || TCNN_MIN_GPU_ARCH <= 52))

to

#define TCNN_HALF_PRECISION 0

Install from a local clone of tiny-cuda-nn, invoke to install Tinycudann to your virtual enviroment

tiny-cuda-nn$ cd bindings/torch
tiny-cuda-nn/bindings/torch$ python setup.py install

3. Visualization Tools

Run Coreset Selection

python coreset.py

TODO

• Coreset selection
• F-Hash input encoding
• Training
• Adaptive Ray Marching (ARM)

Citing H-Hash

If you use it in your research, we would appreciate a citation via

@ARTICLE{sun2025fhash,
  author={Sun, Jianxin and Lenz, David and Yu, Hongfeng and Peterka, Tom},
  journal={IEEE Transactions on Visualization and Computer Graphics}, 
  title={F-Hash: Feature-Based Hash Design for Time-Varying Volume Visualization via Multi-Resolution Tesseract Encoding}, 
  year={2026},
  volume={32},
  number={1},
  pages={396-406},
  keywords={Encoding;Data visualization;Training;Convergence;Rendering (computer graphics);Data models;Superresolution;Hash functions;Computational modeling;Neural radiance field;Time-varying volume;volume visualization;input encoding;deep learning},
  doi={10.1109/TVCG.2025.3634812}
}

License

F-Hash is distributed under the terms of the BSD-3 license.