Kaloian Petkov
ABSTRACT
Neural Radiance Fields (NeRF) trained on pre-rendered photorealistic images represent complex medical data in a fraction of the size, while interactive applications synthesize novel views directly from the neural networks. We demonstrate a practical implementation of NeRFs for high resolution CT volume data, using differentiable rendering for training view selection.
Supplemental Material
Available for Download
- Thomas Kroes, Frits H. Post, and Charl P. Botha. 2012. Exposure Render: An Interactive Photo-Realistic Volume Rendering Framework. PLOS ONE 7, 7 (2012), 1–10. https://doi.org/10.1371/journal.pone.0038586Google Scholar
Cross Ref
- Matthew M. Loper and Michael J. Black. 2014. OpenDR: An Approximate Differentiable Renderer. In Proceedings of ECCV. 154–169.Google Scholar
- Thomas Müller, Alex Evans, Christoph Schied, and Alexander Keller. 2022. Instant Neural Graphics Primitives with a Multiresolution Hash Encoding. ACM Transactions on Graphics 41, 4, Article 102 (2022), 15 pages. https://doi.org/10.1145/3528223.3530127Google ScholarDigital Library
- Paul Tafforeau, Claire Walsh, Willi L. Wagner, Daniyal J. Jafree, Alexandre Bellier, Christopher Werlein, Mark P. Kühnel, Elodie Boller, Simon Walker-Samuel, Jan Lukas Robertus, David A. Long, Joseph Jacob, Sebastian Marussi, Eeline Brown, Natalie Holroyd, Danny D. Jonigk, Maximilian Ackermann, and Peter D. Lee. 2021. Complete brain from the body donor LADAF-2020-31 (Version 1) [Data set]. European Synchrotron Radiation Facility (2021). https://doi.org/doi.org/10.15151/ESRF-DC-572252655Google Scholar
- Sebastian Weiss and Rüdiger Westermann. 2022. Differentiable Direct Volume Rendering. IEEE Transactions on Visualization and Computer Graphics 28, 1 (2022), 562–572. https://doi.org/10.1109/TVCG.2021.3114769Google ScholarDigital Library
- Yiheng Xie, Towaki Takikawa, Shunsuke Saito, Or Litany, Shiqin Yan, Numair Khan, Federico Tombari, James Tompkin, Vincent sitzmann, and Srinath Sridhar. 2022. Neural Fields in Visual Computing and Beyond. Computer Graphics Forum 41, 2 (2022), 641–676. https://doi.org/10.1111/cgf.14505Google ScholarCross Ref
Recommendations
MCNeRF: Monte Carlo Rendering and Denoising for Real-Time NeRFs
SA '23: SIGGRAPH Asia 2023 Conference PapersThe volume rendering step used in Neural Radiance Fields (NeRFs) produces highly photorealistic results, but is inherently slow because it evaluates an MLP at a large number of sample points per ray. Previous work has addressed this by either proposing ...
Reconstructing Translucent Objects using Differentiable Rendering
SIGGRAPH '22: ACM SIGGRAPH 2022 Conference ProceedingsInverse rendering is a powerful approach to modeling objects from photographs, and we extend previous techniques to handle translucent materials that exhibit subsurface scattering. Representing translucency using a heterogeneous bidirectional ...
Comments