An Approximate Graph Elicits Detonation Lattice

Vansh Sharma; Venkat Raman

REVIEW 3 major objections 3 minor 24 references

An Approximate Graph Elicits Detonation Lattice

T0 review · 3 major / 3 minor · reviewed 2026-07-14 · grok-4.5

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A training-free graph algorithm extracts and measures 3D detonation cells from pressure traces.

desk verdict A concrete first-pass automatic 3D detonation-cell pipeline (SAM + graph); useful if the unshown validation holds, and the authors already flag the hard multi-mode cases. read the letter →

arxiv 2603.16524 v2 pith:BDMSBBVT submitted 2026-03-17 cs.CV cs.LGphysics.comp-phphysics.data-an

Vansh Sharma , Venkat Raman This is my paper · ORCID

classification cs.CVcs.LGphysics.comp-phphysics.data-an

keywords 3DdetonationsootfoilgraphcellclassificationcellularlatticeSAMmodelsegmentation

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

Detonation waves leave cellular patterns that researchers still largely measure by hand or with fragile 2D edge detectors. This paper presents a training-free pipeline that first segments 3D pressure fields with a foundation segmentation model, then builds an approximate graph so that each detonation cell can be isolated and measured as a geometric object. On synthetic test volumes the segmentation step errs by about 2 percent; on 3D simulation data the recovered cells are oblong, aligned with the wave direction, and show roughly 17 percent linear size variation whose cubic amplification produces larger scatter in volume. The authors argue that the graph formulation generalizes across cellular geometries and therefore supplies a practical, automatic route to cell statistics that soot-foil methods cannot give in three dimensions. If the method holds up on more irregular multi-mode fields, quantitative 3D cell geometry becomes routine input for detonation-engine design and for studies of triple-point collisions.

What carries the argument

The approximate graph of the detonation lattice: after the segmentation model labels cellular regions, a graph is built whose nodes and edges encode cell connectivity and geometry, allowing each cell to be isolated, sized, and classified without task-specific training.

What would settle it

Run the full pipeline on a 3D detonation simulation or experiment whose multi-mode cellular pattern is known by independent means (for example high-resolution chemiluminescence or reconstructed soot-foil volumes); if measured cell lengths, widths, and volumes systematically disagree with the independent sizes once complexity rises, the claim that the method is a practical general tool fails.

Watch

Extended reading notes

Core claim

A training-free workflow that pairs a foundation segmentation model with an approximate graph construction can automatically detect, segment, and measure three-dimensional detonation cells ("detonation lattices") from volumetric pressure traces, recovering oblong cells aligned with the propagation axis at roughly 17 percent linear deviation and furnishing the first generalized quantitative 3D cell statistics without manual counting or 2D edge heuristics.

Load-bearing premise

That the foundation-model segmentation and the subsequent graph remain faithful on real, highly irregular multi-mode 3D pressure fields so the extracted sizes and volumes are physical measurements rather than segmentation artifacts.

Editorial extensions

If this is right

Three-dimensional cell length, width, and volume statistics can be obtained automatically from simulation pressure fields without manual counting.
Oblong alignment of cells with the wave-propagation axis becomes a quantifiable, reproducible geometric signature rather than a qualitative observation.
The same graph representation supplies a natural data structure for tracking triple-point collisions and their effect on local cell size.
Designers of rotating, pulse, or oblique detonation engines gain a route to 3D cell metrics that soot-foil methods cannot supply.
Comparative studies across fuels and confinements can treat cell-volume distributions as standard outputs instead of rare, labor-intensive measurements.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the graph nodes can be tracked through successive time slices, the same pipeline could turn into a Lagrangian description of how individual cells form, merge, and die.
The cubic amplification of linear size scatter into volume scatter implies that any uncertainty in the segmentation boundary will be magnified in volume-based stability criteria; future work may need boundary-aware uncertainty estimates.
Because the method is training-free, it can be applied immediately to existing large simulation archives without re-labeling campaigns, lowering the barrier for community-wide 3D cell databases.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Punchline: this is a methods paper that claims the first training-free, graph-based pipeline for segmenting and measuring 3D detonation cells from pressure fields. If the full validation holds, people who still count cells by hand or from 2D soot foils will care.

What is actually new is the combination. They run a foundation segmenter (SAM-style) on volumetric pressure traces, then build a graph so the cells become measurable objects—lengths, volumes, joint densities—rather than 2D edge sketches. That sits past the 2D CV/ML and autocorrelation work they cite (including their own soot-foil paper) and past Monnier’s graph/geometric width analyses, which did not deliver a full automatic 3D cell-measurement workflow. Headline numbers: ~2% segmentation error on two synthetic sets; on 3D simulation data, oblong cells aligned with the wave at about 17% linear scatter, with larger volume dispersion from cubic amplification. They also state plainly that highly complex cellular patterns remain hard to segment. That honesty is a plus.

The soft spot is the one they name themselves: faithfulness on irregular multi-mode fields. The text we have is abstract, novelty blurb, and references—no algorithm steps, graph construction from masks, ground-truth definition for the 2%, baselines, or failure-mode tables. Free parameters (prompts, post-processing thresholds) are unspecified here. So the “practical tool / generalizes across geometries” claim rests on the synthetic result and whatever the simulation cases show in the full manuscript. That is a methods soft spot, not a conceptual collapse; circularity burden is low.

Who it is for: combustion and propulsion people who need cell statistics from 3D CFD or experiments, plus CV people doing scientific volumetric segmentation. Citation pattern looks normal for the subfield. Math and data cannot be audited from what is here, but the framing is coherent and the gap is real.

I would send it to peer review rather than desk-reject. Engage if you work on cellular detonations or scientific segmentation; otherwise the headline stats are enough.

Referee Report

3 major / 3 minor

Summary. The manuscript proposes a training-free algorithm that combines a foundation segmentation model with a graph-theoretic workflow to automatically detect, segment, and measure three-dimensional detonation cells from volumetric pressure fields (termed detonation lattices). On two synthetic datasets the segmentation stage is reported to achieve about 2% error; on 3D simulation data the extracted cells are described as oblong and aligned with the wave-propagation axis, with about 17% linear size deviation and larger volume dispersion attributed to cubic amplification of linear variability. The authors position the method as a practical tool for detonation analysis and a foundation for future triple-point collision studies, while acknowledging remaining difficulty with highly complex cellular patterns.

Significance. Automated quantitative 3D measurement of detonation cells would address a genuine methodological gap: the field still relies largely on manual counting or 2D edge heuristics. A training-free, graph-based formulation that generalizes across cellular geometries would be of clear practical value for rotating, pulse, and oblique detonation engines and for fundamental studies of cellular instability. The novelty claim of a first generalized 3D algorithm is plausible relative to the cited 2D and limited 3D literature. Significance cannot yet be fully credited, however, because the algorithm, validation protocol, free parameters, and failure-mode analysis are not present in the submitted text for scrutiny.

major comments (3)

[Manuscript body (missing Methods/Results)] The submitted manuscript contains only the title page, abstract, novelty statement, keywords, and reference list. There is no Methods section, algorithm description, equations, pseudocode, figures, tables, or detailed Results. Load-bearing claims—including how the graph is built from segmentation masks, the definition of the 2% synthetic error, the protocol behind the 17% linear-deviation statistics, and any quantitative evidence of generalization—cannot be verified. A complete technical body is required before the central claims can be assessed.
[Abstract] The abstract simultaneously asserts that the framework is robust and generalizes across diverse cellular geometries (positioning it as a practical tool) and that it remains challenging to reliably segment highly complex cellular patterns. Without quantitative failure-mode rates on multi-mode irregular lattices, success/failure criteria, or comparison against manual or prior 2D baselines, this tension leaves the practical-tool and generalization claims unsupported.
[Abstract / validation claims] The reported 2% segmentation error on synthetic data and 17% linear deviation (with cubic volume scatter) on simulation data are the central quantitative results, yet the available text gives no error metric definition, no synthetic ground-truth construction, no named foundation model or prompting protocol, and no thresholds or free parameters for graph post-processing. These omissions prevent assessment of whether extracted cell sizes/volumes are physical measurements or segmentation artifacts—the binding assumption of the work.

minor comments (3)

[Abstract] Typographical error: 'detonatin lattices' should read 'detonation lattices'.
[Title / Abstract] Title and abstract introduce 'detonation lattice' as the operational object without a precise definition in the available text; once the body is restored, define the lattice (nodes/edges, relation to triple points and cell faces) early and consistently.
[References] References include relevant prior soot-foil ML, 3D cell analyses, and SAM/Cellpose-SAM work. When the body is restored, map in-text citations clearly to the claimed baselines and to the specific foundation segmentation model used.

Circularity Check

0 steps flagged · score 0.0 of 10

Methods paper with empirical segmentation metrics; no derivation chain that reduces predictions to inputs by construction.

full rationale

The available manuscript (title, abstract, novelty statement, keywords, references) presents a training-free graph-based workflow on top of a foundation segmentation model for measuring 3D detonation cells, and reports empirical figures (≈2% synthetic segmentation error; ≈17% linear cell-size scatter on simulation data). There is no claimed first-principles derivation, uniqueness theorem, or fitted parameter that is then re-labeled as a prediction. Self-citations ([10], [18], [24]) appear only as related prior tooling and simulation context by the same group; none is invoked as a load-bearing uniqueness or forcing argument that would make the reported 2%/17% figures true by construction. No equations or algorithm steps in the supplied text equate an output quantity to an input fit. Per the circularity criteria, this is an ordinary methods/measurement paper with independent empirical content; score 0 with empty steps is the correct outcome.

Assumptions & free parameters 2 free parameters · 4 assumptions · 1 invented entities

Central claims rest on domain structure of detonation cells, on foundation-model segmentation transferring to pressure lattices without task-specific training, and on a graph abstraction that turns segments into measurable cells. No free physical constants are fitted in the abstract; algorithmic thresholds and model choices are unstated free parameters. 'Detonation lattice' is a naming/framing entity for the 3D pressure cellular field.

free parameters (2)

Segmentation / graph post-processing thresholds (unspecified)
Any training-free SAM-plus-graph pipeline typically needs prompts, confidence cutoffs, connectivity rules, and size filters; none are given numerical values in the available text but the reported 2% and cell statistics depend on them.
Choice of foundation segmentation model and prompts
Abstract says 'using a segmentation model' and keywords list SAM; model variant and prompting strategy are free design choices that control measured cell boundaries.

assumptions (4)

domain assumption 3D detonation pressure fields form discrete cellular regions whose boundaries can be treated as a lattice/graph of measurable cells.
Load-bearing domain premise throughout abstract and novelty statement; without discrete cell topology the graph measurement is ill-defined.
ad hoc to paper A foundation segmentation model can extract those cells without task-specific training (training-free claim).
Core methodological premise; efficacy asserted via 2% synthetic error but not derived.
standard math Linear cell-size variability is cubically amplified into volume dispersion for the reported statistics.
Abstract's volume-dispersion interpretation assumes roughly geometric scaling of volume with linear dimensions.
domain assumption Prior 2D edge heuristics and manual counting are insufficient for robust 3D volumetric cell measurement.
Motivation axiom in abstract/novelty; frames need for the new pipeline.

invented entities (1)

Detonation lattice (as operational object of the algorithm)
purpose: Name and structure the 3D pressure cellular pattern so graph segmentation/measurement can be applied.
Term is introduced in title/abstract as the thing extracted; it is a framing of existing detonation cell structure rather than a new physical particle, but the pipeline treats it as a discrete graph object.

how reviews work

0 comments

Cite this review

Pith. "Pith review of An Approximate Graph Elicits Detonation Lattice." pith.science (2026). https://pith.science/paper/BDMSBBVT

@misc{pith2026260316524,
  author       = {Pith},
  title        = {Pith review of: An Approximate Graph Elicits Detonation Lattice},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BDMSBBVT}},
  note         = {Machine review of arXiv:2603.16524}
}

read the original abstract

This study presents a novel algorithm based on graph theory for the precise segmentation and measurement of detonation cells from 3D pressure traces, termed detonation lattices, addressing the limitations of manual and primitive 2D edge detection methods prevalent in the field. Using a segmentation model, the proposed training-free algorithm is designed to accurately extract cellular patterns, a longstanding challenge in detonations research. First, the efficacy of segmentation phase on two synthetic datasets is evaluated with an error of 2%. Next, 3D simulation data is used to establish performance of the graph-based workflow. The results of statistics and joint probability densities show oblong cells aligned with the wave propagation axis with 17% deviation, whereas larger dispersion in volume reflects cubic amplification of linear variability. Although the framework is robust, it remains challenging to reliably segment and quantify highly complex cellular patterns. However, the graph-based formulation generalizes across diverse cellular geometries, positioning it as a practical tool for detonation analysis and a strong foundation for future extensions in triple-point collision studies.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

24 extracted references · 1 linked inside Pith

[1]

J. A. Nicholls, H. Wilkinson, R. B. Morrison, Inter- mittent detonation as a thrust-producing mechanism, Journal of jet propulsion 27 (5) (1957) 534–541

1957
[2]

A. P. Nair, A. R. Keller, N. Q. Minesi, D. I. Pineda, R. M. Spearrin, Detonation cell size of liquid hyper- golic propellants: Estimation from a non-premixed combustor, Proceedings of the Combustion Institute 39 (3) (2023) 2757–2765

2023
[3]

Carter, D

M. Carter, D. L. Blunck, Direct comparison of schlieren and soot foil measurements of detonation cell sizes, Frontiers in Aerospace Engineering 1 (2022) 892330

2022
[4]

Shepherd, I

J. Shepherd, I. Moen, S. Murray, P. Thibault, Analyses of the cellular structure of detonations, in: Symposium (International) on Combustion, V ol. 21, Elsevier, 1988, pp. 1649–1658

1988
[5]

B. M. Ng, E. N. Hoffman, D. I. Pineda, C. S. Combs, Detonation cell size estimation via chemiluminescence imaging in an optically accessible linear detonation tube, Experiments in Fluids 65 (7) (2024) 108

2024
[6]

Jalontzki, A

D. Jalontzki, A. Zussman, S. Pendurkar, G. Sharon, Y . Kozak, A computer vision approach for analysis of detonation cellular structures, Applications in Energy and Combustion Science (2025) 100340

2025
[7]

J. E. Shepherd, Detonation in gases, Proceedings of the Combustion Institute 32 (1) (2009) 83–98

2009
[8]

J. Lee, D. Garinis, D. Frost, J. Lee, R. Knystautas, Two-dimensional autocorrelation function analysis of smoked foil patterns, Shock Waves 5 (1995) 169–174

1995

Show all 24 references

[9]

Siatkowski, K

S. Siatkowski, K. Wacko, J. Kindracki, Predicting det- onation cell size of biogas–oxygen mixtures using ma- chine learning models, Shock Waves (2024) 1–9

2024
[10]

Sharma, M

V . Sharma, M. Ullman, V . Raman, A machine learn- ing based approach for statistical analysis of detona- tion cells from soot foils, Combustion and Flame 274 (2025) 114026

2025
[11]

Monnier, V

V . Monnier, V . Rodriguez, P. Vidal, R. Zitoun, An analysis of three-dimensional patterns of experimental detonation cells, Combustion and Flame 245 (2022) 112310

2022
[12]

Monnier, P

V . Monnier, P. Vidal, V . Rodriguez, R. Zitoun, From graph theory and geometric probabilities to a repre- sentative width for three-dimensional detonation cells, Combustion and Flame 256 (2023) 112996

2023
[13]

Borisov, A

S. Borisov, A. Kudryavtsev, Investigation of cellular detonation structure formation via linear stability the- ory and 2d and 3d numerical simulations, in: AIP Con- ference Proceedings, V ol. 1893, AIP Publishing LLC, 2017, p. 030042

2017
[14]

D. N. Williams, L. Bauwens, E. S. Oran, Detailed structure and propagation of three-dimensional deto- nations, in: Symposium (International) on Combus- tion, V ol. 26, Elsevier, 1996, pp. 2991–2998

1996
[15]

Crane, J

J. Crane, J. T. Lipkowicz, X. Shi, I. Wlokas, A. M. Kempf, H. Wang, Three-dimensional detonation struc- ture and its response to confinement, Proceedings of the combustion institute 39 (3) (2023) 2915–2923

2023
[16]

Iwata, S

K. Iwata, S. Suzuki, R. Kai, R. Kurose, Direct nu- merical simulation of detonation–turbulence interac- tion in hydrogen/oxygen/argon mixtures with a de- tailed chemistry, Physics of Fluids 35 (4) (2023)

2023
[17]

Tsuboi, S

N. Tsuboi, S. Katoh, A. K. Hayashi, Three- dimensional numerical simulation for hydrogen/air detonation: Rectangular and diagonal structures, Pro- ceedings of the Combustion Institute 29 (2) (2002) 2783–2788

2002
[18]

Abisleiman, V

S. Abisleiman, V . Sharma, R. Bielawski, V . Raman, Structure of three-dimensional conical oblique det- onation waves, Combustion and Flame 274 (2025) 113971

2025
[19]

Sanchez-Gonzalez, J

A. Sanchez-Gonzalez, J. Godwin, T. Pfaff, R. Ying, J. Leskovec, P. Battaglia, Learning to simulate com- plex physics with graph networks, in: International conference on machine learning, PMLR, 2020, pp. 8459–8468

2020
[20]

Kirillov, E

A. Kirillov, E. Mintun, N. Ravi, H. Mao, C. Rolland, L. Gustafson, T. Xiao, S. Whitehead, A. C. Berg, W.- Y . Lo, et al., Segment anything, in: Proceedings of the IEEE/CVF international conference on computer vi- sion, 2023, pp. 4015–4026

2023
[21]

Pachitariu, M

M. Pachitariu, M. Rariden, C. Stringer, Cellpose-sam: superhuman generalization for cellular segmentation, bioRxiv (2025) 2025–04

2025
[22]

N. Ravi, V . Gabeur, Y .-T. Hu, R. Hu, C. Ryali, T. Ma, H. Khedr, R. R ¨adle, C. Rolland, L. Gustafson, et al., Sam 2: Segment anything in images and videos, arXiv preprint arXiv:2408.00714 (2024)

2024 arXiv
[23]

H. S. M. Coxeter, Regular polytopes, Courier Corpo- ration, 1973

1973
[24]

Oladipo, S

A. Oladipo, S. S. Abisleiman, V . Sharma, J. Singh, V . Raman, S. Dwivedi, T. Hedman, Three observations on cell formation in confined detonations, Manuscript under review (2025)

2025

Pith tools

Reviewed July 14, 2026 · model on record in the stance chip above.

[1] [1]

J. A. Nicholls, H. Wilkinson, R. B. Morrison, Inter- mittent detonation as a thrust-producing mechanism, Journal of jet propulsion 27 (5) (1957) 534–541

1957

[2] [2]

A. P. Nair, A. R. Keller, N. Q. Minesi, D. I. Pineda, R. M. Spearrin, Detonation cell size of liquid hyper- golic propellants: Estimation from a non-premixed combustor, Proceedings of the Combustion Institute 39 (3) (2023) 2757–2765

2023

[3] [3]

Carter, D

M. Carter, D. L. Blunck, Direct comparison of schlieren and soot foil measurements of detonation cell sizes, Frontiers in Aerospace Engineering 1 (2022) 892330

2022

[4] [4]

Shepherd, I

J. Shepherd, I. Moen, S. Murray, P. Thibault, Analyses of the cellular structure of detonations, in: Symposium (International) on Combustion, V ol. 21, Elsevier, 1988, pp. 1649–1658

1988

[5] [5]

B. M. Ng, E. N. Hoffman, D. I. Pineda, C. S. Combs, Detonation cell size estimation via chemiluminescence imaging in an optically accessible linear detonation tube, Experiments in Fluids 65 (7) (2024) 108

2024

[6] [6]

Jalontzki, A

D. Jalontzki, A. Zussman, S. Pendurkar, G. Sharon, Y . Kozak, A computer vision approach for analysis of detonation cellular structures, Applications in Energy and Combustion Science (2025) 100340

2025

[7] [7]

J. E. Shepherd, Detonation in gases, Proceedings of the Combustion Institute 32 (1) (2009) 83–98

2009

[8] [8]

J. Lee, D. Garinis, D. Frost, J. Lee, R. Knystautas, Two-dimensional autocorrelation function analysis of smoked foil patterns, Shock Waves 5 (1995) 169–174

1995