Writing study: Neural Field-Based 3D Surface Reconstruction of Microstructures from Multi-Detector Signals in Scanning Electron Microscopy

CVPR overviewCorpus atlasBy contribution typeVisual galleryRules & templateWriting patternsPaper studies

← CVPR 2026 oral writing corpus

Shuo Chen, Yijin Li, Xi Zheng, Guofeng Zhang · Official CVPR 2026 oral record

Observed writing choices, not acceptance rules. This study examines the inspected manuscript version. It does not establish why the paper was accepted or that its presentation choices should be copied unchanged.

Figure, table and section ledger

Editorial contribution type: Domain-grounded methods & physical systems (primary).

Layout extraction, not a required paper template. Counts are distinct caption labels; a multi-panel figure counts once. Section spans are estimated from heading positions, including floats and whitespace. Method and limitations.

Canonical PDF

12 physical pages

Before references / appendix

≈ 8.15 page-equivalents

Main figures / tables

6 / 1

Attached appendix figures / tables

0 / 0

Attached appendix span

Separate file recorded below; not in this PDF total

Before references / appendix: ≈ 67.92%References / intervening material: ≈ 32.08%Estimated layout envelopes, not text/artwork area. Unresolved appendix boundaries remain within the undivided post-body material.

Extraction notes: No attached appendix boundary detected. Zero means no recognized caption, not independently verified absence.

Section-by-section lengths and source pages

The title/abstract/front matter precede the first numbered heading. A span ends at the next detected top-level heading, not at the last sentence; end-page whitespace is included. References and statements are listed separately. These are not measurements of text-only or artwork area.

Top-level heading	Starts on PDF page	Estimated span	Share of entire PDF
1. Introduction	1	≈ 0.94 pp.	7.83%
2. Related Work	2	≈ 0.67 pp.	5.58%
3. Method	3	≈ 2.48 pp.	20.67%
4. Experiments	5	≈ 2.95 pp.	24.58%
5. Conclusion	8	≈ 0.25 pp.	2.08%
Acknowledgements	9	≈ 0.1 pp.	0.83%
References	9	≈ 3.85 pp.	32.08%

All detected figure and table locations

Main text

• Figure 1 · p. 1
• Figure 2 · p. 3
• Figure 3 · p. 4
• Figure 4 · p. 6
• Figure 5 · p. 7
• Figure 6 · p. 8
• Table 1 · p. 8

Attached appendix

No recognized caption labels.

Separate source files—excluded from canonical-PDF totals

A separate author version may repeat main text and references. Whole-file lengths and counts are not extra appendix material and are never added to the canonical PDF totals. Read the study's source scope for which sections received substantive reading versus layout inspection.

Additional source 1

Separate PDF (or ZIP containing it): whole-file layout of 9 physical pages; 8 detected figures and 0 tables. These may include overlapping main-text material.

Located appendix only: starts on physical p.1; ≈ 8.93 page-equivalents, 8 figure labels and 0 table labels. This interval is distinct from the whole file.

Detected heading	PDF page	Estimated span
Supplementary Material	1	≈ 0.02 pp.
1. Multi-View and Multi-Detector SEM Data	1	≈ 0.6 pp.
2. Neural Field Implementation Details	1	≈ 0.36 pp.
3. Sample Preparation	2	≈ 0.31 pp.
4. Challenges in Acquiring Ground-Truth 3D	2	≈ 0.47 pp.
5. Construction of Simulated Dataset	2	≈ 0.55 pp.
6. Evaluation Metrics	3	≈ 0.49 pp.
7. Additional Experiments	3	≈ 5.19 pp.
References	9	≈ 0.95 pp.

• Figure 1 · p. 1
• Figure 2 · p. 3
• Figure 3 · p. 5
• Figure 4 · p. 6
• Figure 5 · p. 7
• Figure 6 · p. 7
• Figure 7 · p. 8
• Figure 8 · p. 8

Argument and organization

1. Problem and contribution

   The paper motivates why multi-detector electron signals should constrain 3D microgeometry, rather than treating microscopy as an ordinary RGB reconstruction task. The physical measurement mechanism is established before the learned model.

2. Argument sequence

   Acquisition and emission preliminaries lead into a hybrid reconstruction pipeline and its losses. Experiments then distinguish captured specimens, simulated cases and component studies, which is essential when dense real ground truth is difficult to obtain.

3. Evidence and verification

   Table 1 reports geometric distance, normal error and detector-signal error separately for ablations. Missing signal-model measurements are shown as unavailable rather than zero, and the geometry columns reveal that removing a term can improve one metric while harming another.

Problem, claim and contribution

The paper motivates why multi-detector electron signals should constrain 3D microgeometry, rather than treating microscopy as an ordinary RGB reconstruction task. The physical measurement mechanism is established before the learned model.

Introduction and preliminaries, Sections 1 and 3 · Main paper · Additional source 2

Argument and section order

Acquisition and emission preliminaries lead into a hybrid reconstruction pipeline and its losses. Experiments then distinguish captured specimens, simulated cases and component studies, which is essential when dense real ground truth is difficult to obtain.

Sections 3–5; supplementary Sections 4–6 · Main paper · Additional source 2

Figures, captions and visual explanation

Figure 2 links the detector setup, coarse reconstruction, neural field, forward BSE model and losses. Figure 6 compares real/simulated signals with intensity profiles and shadow masks, giving readers more diagnostic information than a final mesh alone.

Figure 2, PDF p.3; Figure 6, PDF p.8 · Main paper · Additional source 2

Tables, comparisons and evidence

Table 1 reports geometric distance, normal error and detector-signal error separately for ablations. Missing signal-model measurements are shown as unavailable rather than zero, and the geometry columns reveal that removing a term can improve one metric while harming another.

Table 1, PDF p.8 · Main paper · Additional source 2

Appendix and reproducibility

The supplement explains sample preparation and the difficulty of obtaining trustworthy microstructure ground truth, including why nominal design geometry is not automatically measured truth. It also adds detector and initialization analyses.

Supplementary Sections 1–7, especially ground-truth acquisition and evaluation sections · Main paper · Additional source 2

Clarity and writing conventions

The strongest captions identify whether a result is physical measurement, simulation or reconstructed prediction. This labeling discipline prevents qualitative agreement on real specimens from being presented as dense quantitative validation.

Figures 4–6; supplementary simulation and ground-truth discussion · Main paper · Additional source 2

What to reuse—and what not to infer

Transferable lesson: For scientific-imaging work, explain the measurement physics and ground-truth limitations before presenting reconstruction metrics; show raw-signal agreement alongside geometry.

Boundary: An ablation that is best on one geometric metric need not be best on normals or signal fit. Do not turn the multi-objective table into an unqualified claim that every component improves every score.

Apply this lesson using the argument and evidence worksheet.

Inspection record and sources

Checked

2026-09-26

Reading status

Writing-focused manuscript scan + selected visuals inspected

Main text

Writing-oriented scan throughout main PDF pp.1–8 and references pp.9–12: abstract/framing, section sequence, figure/table captions, evaluation organization and conclusions; not an exhaustive technical verification.

Appendix / supplement

Writing-oriented supplement scan: all nine supplementary PDF pages covering acquisition, sample preparation, implementation, evaluation difficulties, simulations and extra analyses. Video/code assets were not executed or assessed.

Visually inspected PDF pages

3, 8

This is a writing-focused scan of the manuscript and the stated supplementary scope: framing, organization, visual explanation, evidence presentation and wording. It is not a line-by-line proof audit, reproduction of results or exhaustive scientific review.

• Official CVF open-access CVPR 2026 accepted-paper PDF · 12 pages
• Official CVF supplementary ZIP member PDF; see URL fragment for member · 9 pages