arxiv: 2604.02980 · v1 · submitted 2026-04-03 · 💻 cs.HC

Recognition: no theorem link

UnrealVis: A Testing Laboratory of Optimization Techniques in Unreal Engine for Scientific Visualization

Matteo Filosa , Andrea Nardocci , Tiziana Catarci , Marco Angelini

Authors on Pith no claims yet

Pith reviewed 2026-05-13 18:33 UTC · model grok-4.3

classification 💻 cs.HC

keywords scientific visualizationUnreal Engineoptimization techniquesrendering performancelevel of detailtaxonomy3D datasetsinteractive exploration

0 comments

The pith

UnrealVis provides a configurable lab in Unreal Engine to test and select rendering optimizations that reach performance targets while keeping structural fidelity in large scientific 3D datasets.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

Large scientific 3D datasets demand careful choices among rendering techniques to avoid slow frame rates or loss of important visual details. UnrealVis builds an interactive laboratory inside Unreal Engine that draws on a taxonomy of 22 techniques grouped in six families, derived from a survey of 55 papers. The system exposes these methods through engine features such as Nanite and level-of-detail controls, and supplies live telemetry plus side-by-side A/B comparisons so users can measure both local and global effects. Case studies on ribosomal structures and volumetric flow fields, together with expert feedback, demonstrate that the laboratory helps identify workable combinations that satisfy speed requirements without sacrificing key structural elements.

Core claim

UnrealVis establishes a taxonomy of 22 optimization techniques across six families from a review of 55 papers and implements them through Unreal Engine subsystems including Nanite, LOD schemes, and culling. Its workflow of live telemetry and A/B comparisons enables users to evaluate and choose optimization sets that meet performance goals while preserving structural fidelity, as validated on ribosomal and flow-field datasets.

What carries the argument

UnrealVis, the Unreal Engine laboratory that supplies a taxonomy of 22 techniques in six families together with real-time telemetry and A/B comparison tools for configuring rendering subsystems.

Load-bearing premise

The taxonomy drawn from 55 papers and the implemented Unreal Engine subsystems cover the relevant techniques, and results from the ribosomal and flow-field studies generalize to other scientific datasets.

What would settle it

Finding a new scientific dataset for which no combination of the 22 available optimizations meets both the target frame rate and acceptable structural fidelity would show that UnrealVis does not facilitate such selections.

Figures

Figures reproduced from arXiv: 2604.02980 by Andrea Nardocci, Marco Angelini, Matteo Filosa, Tiziana Catarci.

**Figure 2.** Figure 2: The UnrealVis optimization taxonomy: categories and specific [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** Figure 3: Technical schematic of the UnrealVis data ingestion pipeline, [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

**Figure 4.** Figure 4: An example workflow for starting a simulation in UnrealVis, showing a huge bacterial 70S ribosome. The user is first greeted by a welcome [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗

**Figure 6.** Figure 6: BLASTNet 2.0 volumetric exploration. The main view tracks [PITH_FULL_IMAGE:figures/full_fig_p007_6.png] view at source ↗

**Figure 7.** Figure 7: Aggregated Likert scale responses (N = 4). High scores in Tasks 1–2 confirm the system’s ease of use for general navigation. The score decrease in Tasks 4–5 (analysis and comparison) highlights the limitations of human perception in quantifying optimization impacts during manual interaction, justifying the need for deterministic benchmarks. 5.3 Quantitative Benchmarking To address the trajectory inconsiste… view at source ↗

read the original abstract

Visualizing large 3D scientific datasets requires balancing performance and fidelity, but traditional tools often demand excessive technical expertise. We introduce UnrealVis, an Unreal Engine optimization laboratory for configuring and evaluating rendering techniques during interactive exploration. Following a review of 55 papers, we established a taxonomy of 22 optimization techniques across six families, implementing them through engine subsystems such as Nanite, Level of Detail(LOD) schemes, and culling. The system features an intuitive workflow with live telemetry and A/B comparisons for local and global performance analysis. Validated through case studies of ribosomal structures and volumetric flow fields, along with an expert evaluation, UnrealVis facilitates the selection of optimization combinations that meet performance goals while preserving structural fidelity. UnrealVis is available at https://github.com/XAIber-lab/UnrealVis

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

UnrealVis gives a concrete Unreal Engine lab for testing rendering optimizations on scientific data plus a taxonomy of 22 techniques, but the validation stays narrow and lacks numbers.

read the letter

UnrealVis is a testing lab built inside Unreal Engine that lets users configure and compare rendering optimizations for large 3D scientific datasets. The authors reviewed 55 papers, extracted a taxonomy of 22 techniques across six families, and wired them into engine features such as Nanite, LOD schemes, and culling. They added a workflow with live telemetry and A/B comparisons so people can check local and global performance on the fly. The code is on GitHub, which makes the whole thing usable right away. The ribosomal structure and volumetric flow-field case studies plus the expert evaluation show the system in action on real data. That part is straightforward engineering work that organizes existing tools into something accessible for non-experts. The soft spot is the validation. The paper claims the combinations preserve structural fidelity while hitting performance targets, yet it gives no quantitative metrics, error bars, or detailed methodology on how fidelity was measured. The two datasets share traits like discrete geometry and coherent fields, so it is not clear whether the same combinations hold up for unstructured meshes, time-varying medical volumes, or sparse point clouds. Without cross-dataset checks or sensitivity analysis, the facilitation claim stays scoped to the tested instances. This paper is for visualization researchers and developers who already work with Unreal or want a ready-made way to tune large-data rendering. A practitioner who needs to balance speed and accuracy on similar data would get immediate value from the implementation and the organized list of techniques. It deserves peer review because the system is shipped and open, even though reviewers will need to press for stronger quantitative evidence and broader testing.

Referee Report

2 major / 3 minor

Summary. The manuscript introduces UnrealVis, an Unreal Engine-based laboratory for configuring, testing, and evaluating optimization techniques for interactive visualization of large 3D scientific datasets. Following a review of 55 papers, the authors derive a taxonomy of 22 techniques across six families and implement them via engine subsystems such as Nanite, LOD schemes, and culling. The system offers an intuitive workflow with live telemetry and A/B comparisons for local and global performance analysis. Validation is performed through case studies on ribosomal structures and volumetric flow fields together with an expert evaluation. The central claim is that UnrealVis enables users to select optimization combinations that satisfy performance targets while preserving structural fidelity.

Significance. If the claims hold, UnrealVis would provide a practical, accessible platform that lowers the expertise barrier for domain scientists to leverage high-performance game-engine rendering in their work. The open-source release, structured taxonomy, and focus on fidelity-preserving trade-offs address a recurring challenge in scientific visualization. The work could serve as a reusable testbed for future studies on interactive rendering optimizations.

major comments (2)

[Case Studies] Case Studies section: Validation rests on only two narrow datasets (ribosomal structures and volumetric flow fields) plus expert evaluation. These share specific traits (discrete geometry, coherent fields) that may not represent other domains such as unstructured meshes or time-varying medical volumes. No cross-dataset validation or sensitivity analysis is reported, so the claim that selected combinations preserve fidelity in general is not load-bearing supported.
[Validation] Validation section: The manuscript states that validation occurred through case studies and expert evaluation yet supplies no quantitative metrics, error analysis, fidelity measures (e.g., structural similarity, visual quality scores), or detailed methodology for assessing preservation of structural fidelity. This absence directly undermines verification of the central facilitation claim.

minor comments (3)

[Abstract] Abstract: Adding one or two concrete performance numbers or fidelity observations from the case studies would strengthen the summary of results.
[Implementation] Implementation: A table mapping the six taxonomy families to the specific Unreal Engine subsystems (Nanite, LOD, culling) would improve clarity and traceability.
[Workflow] Workflow description: Additional figures or annotated screenshots illustrating the live telemetry and A/B comparison interface would aid reader comprehension of the claimed intuitive workflow.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive review. The comments highlight important limitations in the current validation strategy. We agree that the evidence for generalizability and quantitative fidelity assessment requires strengthening and will revise the manuscript accordingly.

read point-by-point responses

Referee: [Case Studies] Case Studies section: Validation rests on only two narrow datasets (ribosomal structures and volumetric flow fields) plus expert evaluation. These share specific traits (discrete geometry, coherent fields) that may not represent other domains such as unstructured meshes or time-varying medical volumes. No cross-dataset validation or sensitivity analysis is reported, so the claim that selected combinations preserve fidelity in general is not load-bearing supported.

Authors: We agree that the two datasets share structural characteristics and that broader validation is needed to support claims of general applicability. In the revised manuscript we will add a third case study using an unstructured tetrahedral mesh from computational fluid dynamics and include a sensitivity analysis that varies data resolution, temporal coherence, and mesh irregularity. We will also report cross-dataset performance and fidelity trends to better substantiate the generalizability of the selected optimization combinations. revision: yes
Referee: [Validation] Validation section: The manuscript states that validation occurred through case studies and expert evaluation yet supplies no quantitative metrics, error analysis, fidelity measures (e.g., structural similarity, visual quality scores), or detailed methodology for assessing preservation of structural fidelity. This absence directly undermines verification of the central facilitation claim.

Authors: We acknowledge the absence of quantitative fidelity metrics and detailed methodology in the current draft. The revised version will add a dedicated subsection under Validation that reports SSIM, PSNR, and perceptual quality scores computed against ground-truth high-fidelity renders. We will also describe the exact protocol used for the expert evaluation (including rating scales, number of participants, and statistical analysis) together with error analysis comparing optimized versus baseline renderings across the case studies. revision: yes

Circularity Check

0 steps flagged

No significant circularity; contribution is implementation and taxonomy without derivations

full rationale

The paper introduces a software laboratory (UnrealVis) and a taxonomy of 22 techniques derived from a review of 55 external papers. No equations, fitted parameters, predictions, or mathematical derivations are present. Claims rest on case studies (ribosomal structures, flow fields) and expert evaluation using publicly available Unreal Engine subsystems (Nanite, LOD, culling). The taxonomy is constructed from cited literature rather than self-definition, and no self-citation chain or ansatz is load-bearing for the core facilitation claim. The work is self-contained as an engineering artifact.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The paper rests on a literature review to create the taxonomy and on standard Unreal Engine subsystems (Nanite, LOD, culling) for implementation; no free parameters, new axioms, or invented entities are introduced beyond configuration choices.

axioms (1)

domain assumption Unreal Engine subsystems such as Nanite, LOD schemes, and culling can be configured to support scientific visualization workflows
Invoked when describing the implementation through engine subsystems.

pith-pipeline@v0.9.0 · 5440 in / 1204 out tokens · 41218 ms · 2026-05-13T18:33:34.025119+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

54 extracted references · 54 canonical work pages

[1]

Ayachit.The ParaView Guide: A Parallel Visualization Application

U. Ayachit.The ParaView Guide: A Parallel Visualization Application. Kitware, 2015. 1, 2, 3, 4

work page 2015
[2]

Baaden, O

M. Baaden, O. Delalande, et al. Visualizing biomolecular electrostatics in virtual reality with unitymol-apbs.Faraday Discussions, 209:415–432,

work page
[3]

doi: 10.1039/C8FD00025D 2

work page doi:10.1039/c8fd00025d
[4]

Baaden, M

M. Baaden, M. O’Connor, et al. State of the art of molecular visualization in immersive virtual environments.Computer Graphics Forum, 42(1):e14738,

work page
[5]

doi: 10.1111/cgf.14738 2

work page doi:10.1111/cgf.14738
[6]

Beckmann et al

R. Beckmann et al. Cryo-em structure of human 80s ribosome with mrna and trnas. https://www.rcsb.org/structure/4V6W, 2015. Protein Data Bank entry 4V6W. 4, 5, 9

work page 2015
[7]

H. M. Berman, J. Westbrook, Z. Feng, G. Gilliland, T. N. Bhat, H. Weissig, I. N. Shindyalov, and P. E. Bourne. The protein data bank.Nucleic Acids Research, 28(1):235–242, 2000. doi: 10.1093/nar/28.1.235 1, 5

work page doi:10.1093/nar/28.1.235 2000
[8]

quick and dirty

J. Brooke. Sus: A “quick and dirty” usability scale. In P. W. Jordan, B. Thomas, B. A. Weerdmeester, and I. L. McClelland, eds.,Usability Evaluation in Industry, pp. 189–194. Taylor & Francis, 1996. 6, 7

work page 1996
[9]

Chavent, A

M. Chavent, A. Vanel, A. Tek, B. Levy, S. Robert, B. Raffin, and M. Baaden. GPU-accelerated atom and dynamic bond visualization using HyperBalls: a unified algorithm for balls, sticks, and hyperboloids.Journal of Compu- tational Chemistry, 32(13):2924–2935, 2011. doi: 10.1002/jcc.21861 2, 3

work page doi:10.1002/jcc.21861 2011
[10]

Science and Innovation

J. W. Choi, K. Y . Lee, and O. Narzulloev. Acceleration chimera hologram rendering with unreal engine.International Scientific Journal "Science and Innovation", 3(1), 2024. Series A. 2, 3, 5

work page 2024
[11]

K. W. Chong, V . Krishnan, L. Ho, et al. Gears: A game-engine-assisted research platform for scientific computing.IEEE Computer Graphics and Applications, 37(5):66–79, 2017. doi: 10.1109/MCG.2017.3273307 3, 4, 9

work page doi:10.1109/mcg.2017.3273307 2017
[12]

De Fabrizio et al

R. De Fabrizio et al. Immersive virtual reality in computational chemistry. International Journal of Quantum Chemistry, 116(10):739–749, 2016. doi: 10.1002/qua.25102 2

work page doi:10.1002/qua.25102 2016
[13]

W. L. DeLano. Pymol: An open-source molecular graphics tool.CCP4 Newsletter on Protein Crystallography, 40(1):82–92, 2002. 2

work page 2002
[14]

Drouhard, C

M. Drouhard, C. A. Steed, S. Hahn, J. Daniel, and M. A. Matheson. Immersive visualization for materials science data analysis using the Oculus Rift. In2015 IEEE International Conference on Big Data (Big Data), pp. 2453–2461. IEEE, 2015. doi: 10.1109/BigData.2015.7364040 1

work page doi:10.1109/bigdata.2015.7364040 2015
[15]

M. D. Díaz-Alemán, E. M. Amador-García, E. Díaz-González, and J. de la Torre-Cantero. Nanite as a disruptive technology for the interactive visualisation of cultural heritage 3D models: A case study.Heritage, 6(8):5607–5618, 2023. doi: 10.3390/heritage6080295 3, 4

work page doi:10.3390/heritage6080295 2023
[16]

Lumen global illumination and reflections in Un- real Engine 5

Epic Games. Lumen global illumination and reflections in Un- real Engine 5. https://docs.unrealengine.com/5.0/en-US/ lumen-global-illumination-and-reflections-in-unreal-engine/ ,

work page
[17]

Accessed: 2026-02-04. 2, 3

work page 2026
[18]

Nanite virtualized geometry in Unreal En- gine 5

Epic Games. Nanite virtualized geometry in Unreal En- gine 5. https://docs.unrealengine.com/5.0/en-US/ nanite-virtualized-geometry-in-unreal-engine/ , 2021. Accessed: 2026-02-04. 2, 3, 4

work page 2021
[19]

K. A. Ericsson and H. A. Simon.Protocol Analysis: Verbal Reports as Data. MIT Press, Cambridge, MA, rev. ed. ed., 1993. 6, 7

work page 1993
[20]

Fedotova, M

N. Fedotova, M. Protsenko, I. Baranova, S. Vashchenko, and Y . Dehtiarenko. Research on calculation optimization methods used in computer games development.Infor- matyka, Automatyka, Pomiary w Gospodarce i Ochronie´Srodowiska, 13(3):37–42,

work page
[21]

doi: 10.35784/iapgos.3828 2, 3, 5

work page doi:10.35784/iapgos.3828
[22]

Ferreira, J

C. Ferreira, J. Gomes, I. Matos, A. Gomes, A. C. Paiva, and A. Raposo. GPU shaders and hyperballs for real-time molecular visualization in unity. In2018 IEEE Conference on Virtual Reality and 3D User Interfaces (VR), pp. 655–656, 2018. doi: 10.1109/VR.2018.8446523 3

work page doi:10.1109/vr.2018.8446523 2018
[23]

Fischer et al

N. Fischer et al. Translating 70s ribosome in the unrotated state. https: //www.rcsb.org/structure/8B0X, 2022. Protein Data Bank entry 8B0X. 4, 5, 7, 9

work page 2022
[24]

S. Frey, F. Sadlo, and D. Weiskopf. On evaluating runtime performance of interactive visualizations.IEEE Transactions on Visualization and Computer Graphics, 25(1):626–636, 2019. doi: 10.1109/TVCG.2018.2865233 2, 4, 9

work page doi:10.1109/tvcg.2018.2865233 2019
[25]

S. A. Fromm, F. Weis, J. Reichert, K. H. Nierhaus, and N. Fischer. The translating bacterial ribosome at 1.55 Å resolution generated by cryo-em imaging services. Nature Communications, 14:1095, 2023. doi: 10.1038/s41467-023-36742-3 5

work page doi:10.1038/s41467-023-36742-3 2023
[26]

R. J. García-Hernández and D. Kranzlmüller. Virtual reality toolset for material science: Nomad vr tools. InProceedings of the International Conference on High Performance Computing & Simulation (HPCS), pp. 593–600, 2017. doi: 10.1109/HPCS.2017.104 2

work page doi:10.1109/hpcs.2017.104 2017
[27]

Garland and P

M. Garland and P. S. Heckbert. Surface simplification using quadric error metrics. InProceedings of the 24th annual conference on Computer graphics and interactive techniques (SIGGRAPH), pp. 209–216, 1997. doi: 10.1145/258734.258849 3

work page doi:10.1145/258734.258849 1997
[28]

M. S. Greenwood and M. B. R. Smith. Method for visualizing radiation data using Unreal Engine. InTransactions of the American Nuclear Society, vol. 123 of Transactions of the American Nuclear Society, pp. 1167–1170. American Nuclear Society, La Grange Park, IL, USA, Nov. 2020. Presented at the 2020 American Nuclear Society Winter Meeting (Virtual). doi: 1...

work page doi:10.13182/t123-33177 2020
[29]

Humphrey, A

W. Humphrey, A. Dalke, and K. Schulten. Vmd: Visual molecular dynamics.Journal of Molecular Graphics, 14(1):33–38, 1996. doi: 10.1016/0263-7855(96)00018-5 2, 4

work page doi:10.1016/0263-7855(96)00018-5 1996
[30]

Karis et al

B. Karis et al. Real-time global illumination and shadows with unreal engine 5. In ACM SIGGRAPH Talks, 2021. Virtual Shadow Maps technical overview. 3

work page 2021
[31]

O. Kreylos. Chisel: A next-generation visualization system. InAGU Fall Meeting Abstracts, 2012. Uses Unreal Engine for seismic visualization. 2

work page 2012
[32]

Krüger and R

J. Krüger and R. Westermann. Acceleration techniques for gpu-based volume rendering. InProceedings of the 14th IEEE Visualization Conference (VIS), pp. 287–292, 2003. doi: 10.1109/VISUAL.2003.1250384 3

work page doi:10.1109/visual.2003.1250384 2003
[33]

Lanrezac, N

A. Lanrezac, N. Férey, and M. Baaden. Wielding the power of interactive molecular simulations.WIREs Computational Molecular Science, 12(4):e1594, 2022. doi: 10 .1002/wcms.1594 1, 3, 9

work page 2022
[34]

Laureanti, J

J. Laureanti, J. Brandi, E. Offor, D. Engel, R. Rallo, B. Ginovska, X. Martinez, M. Baaden, and N. A. Baker. Visualizing biomolecular electrostatics in virtual reality with UnityMol-APBS.Protein Science, 29(1):237–246, 2020. doi: 10.1002/pro.3773 2, 3

work page doi:10.1002/pro.3773 2020
[35]

Le Muzic, L

M. Le Muzic, L. Autin, J. Parulek, and I. Viola. cellVIEW: a tool for illustrative and multi-scale rendering of large biomolecular datasets. InEurographics Workshop on Visual Computing for Biology and Medicine (VCBM). The Eurographics Association,

work page
[36]

doi: 10.2312/vcbm.20151209 2, 3

work page doi:10.2312/vcbm.20151209
[37]

Li and K

W. Li and K. Mueller. Empty space skipping and occlusion clipping for texture-based volume rendering. InProceedings of IEEE Visualization (VIS), pp. 317–324, 2003. doi: 10.1109/VISUAL.2003.1250395 3

work page doi:10.1109/visual.2003.1250395 2003
[38]

Y . Li, Z. Wang, T. Wang, H. Wang, W. T. Chung, A. Jagannath, N. Jain, et al. Tur- bulence in focus: Benchmarking scaling behavior of 3D volumetric super-resolution with BLASTNet 2.0. https://arxiv.org/abs/2309.13457, 2023. NeurIPS Datasets & Benchmarks Track. 4, 5, 9

work page arXiv 2023
[39]

W. E. Lorensen and H. E. Cline. Marching cubes: A high resolution 3d surface construction algorithm.ACM SIGGRAPH Computer Graphics, 21(4):163–169, 1987. doi: 10.1145/37402.37422 3

work page doi:10.1145/37402.37422 1987
[40]

Z. Lv, A. Tek, F. Franco, C. Sevilla, P. Derreumaux, B. J. Brewer, and M. Baaden. Game engines for molecular dynamics visualization.PLOS ONE, 8(3):e57418, 2013. doi: 10.1371/journal.pone.0057418 2, 3

work page doi:10.1371/journal.pone.0057418 2013
[41]

Löffler, F

F. Löffler, F. Servén, et al. NOMAD VR: Immersive visualization for materials science.Computer Physics Communications, 246:106864, 2019. doi: 10.1016/j.cpc .2019.106864 2, 3

work page doi:10.1016/j.cpc 2019
[42]

Marsden and F

C. Marsden and F. Shankar. Using unreal engine to visualize a cosmological volume.Universe, 6(10):207, 2020. Preprint arXiv:2010.02942. doi: 10. 3390/universe6100207 2, 4

work page arXiv 2020
[43]

Martinez, M

X. Martinez, M. Krone, and M. Baaden. QuickSES: A library for fast computation of solvent excluded surfaces. InEurographics Workshop on Molecular Graphics and Visual Analysis of Molecular Data (MolVa). The Eurographics Association, 2019. doi: 10.2312/molva.20191095 3

work page doi:10.2312/molva.20191095 2019
[44]

G. Meng, N. Spahich, R. Kenjale, G. Waksman, and J. W. St Geme. Crystal structure of theHaemophilus influenzaehap adhesin. https: //www.rcsb.org/structure/3SYJ, 2011. Protein Data Bank entry 3SYJ. Resolution: 2.20 Å. doi: 10.2210/pdb3SYJ/pdb 1, 5, 7, 9

work page doi:10.2210/pdb3syj/pdb 2011
[45]

G. Meng, N. Spahich, R. Kenjale, G. Waksman, and J. W. St Geme. Crystal structure of theHaemophilus influenzaehap adhesin reveals an intercellular oligomerization mechanism for bacterial aggregation.The EMBO Journal, 30(19):3864–3874, 2011. doi: 10.1038/emboj.2011.279 1, 5

work page doi:10.1038/emboj.2011.279 2011
[46]

Reina, H

G. Reina, H. Childs, K. Matkovi´c, K. Bühler, M. Waldner, D. Pugmire, B. Kozlíková, T. Ropinski, P. Ljung, T. Itoh, E. Gröller, and M. Krone. The moving target of visualization software for an increasingly complex world.Computers & Graphics, 87:12–29, 2020. doi: 10.1016/j.cag.2020.01.005 1, 3, 4, 9

work page doi:10.1016/j.cag.2020.01.005 2020
[47]

Rossant et al

C. Rossant et al. Hardware-accelerated interactive data visualization for neuroscience. Frontiers in Neuroinformatics, 7:36, 2013. doi: 10.3389/fninf.2013.00036 2, 3, 9

work page doi:10.3389/fninf.2013.00036 2013
[48]

M. W. Sawicki, M. Erman, T. Puranen, P. T. Riikonen, and P. Vihko. Structure of the ternary complex of human 17β-hydroxysteroid dehydrogenase type 1 with equilin and nadp+.Proceedings of the National Academy of Sciences, 96(3):840–845, 1999. doi: 10.1073/pnas.96.3.840 5, 6

work page doi:10.1073/pnas.96.3.840 1999
[49]

M. W. Sawicki, M. Erman, T. Puranen, P. T. Riikonen, and P. Vihko. Type 1 17-beta hydroxysteroid dehydrogenase equilin complexed with nadp+. https://www.rcsb.org/structure/1EQU, 1999. Protein Data Bank entry 1EQU. Resolution: 2.20 Å. Release: 1999-12-02. doi: 10.2210/pdb1EQU/pdb 5, 6, 7, 9

work page doi:10.2210/pdb1equ/pdb 1999
[50]

T. Ulrich. Rendering massive terrains using chunked level of detail control. In SIGGRAPH Course Notes, 2002. 3

work page 2002
[51]

I. Wald, G. P. Johnson, J. Amstutz, C. Brownlee, A. Knoll, J. Jeffers, J. Günther, and P. A. Navrátil. OSPRay: A CPU ray tracing framework for scientific visualization. InIEEE Transactions on Visualization and Computer Graphics, vol. 23, pp. 931–940,

work page
[52]

doi: 10.1109/TVCG.2016.2617865 1, 2, 3

Presented at IEEE VIS 2016. doi: 10.1109/TVCG.2016.2617865 1, 2, 3

work page doi:10.1109/tvcg.2016.2617865 2016
[53]

J. Wang, N. Xiang, N. Kukreja, L. Yu, and H.-N. Liang. LVDIF: A framework for real-time interaction with large volume data.The Visual Computer, 39(8):3373–3386,

work page
[54]

doi: 10.1007/s00371-023-02976-x 1, 2, 3

work page doi:10.1007/s00371-023-02976-x