Writing study: TESO: Online Tracking of Essential Matrix by Stochastic Optimization

CVPR overviewCorpus atlasBy contribution typeVisual galleryRules & templateWriting patternsPaper studies

← CVPR 2026 oral writing corpus

Jaroslav Moravec, Radim Sara, Akihiro Sugimoto · 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

9 physical pages

Before references / appendix

≈ 7.76 page-equivalents

Main figures / tables

3 / 4

Attached appendix figures / tables

0 / 0

Attached appendix span

Separate file recorded below; not in this PDF total

Before references / appendix: ≈ 86.22%References / intervening material: ≈ 13.78%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	≈ 1.22 pp.	13.56%
2. Related Work	2	≈ 0.62 pp.	6.89%
3. Methods	3	≈ 1.81 pp.	20.11%
4. Datasets	5	≈ 0.54 pp.	6%
5. Experiments	5	≈ 2.76 pp.	30.67%
6. Conclusion	8	≈ 0.33 pp.	3.67%
Acknowledgments	8	≈ 0.09 pp.	1%
References	8	≈ 1.24 pp.	13.78%

All detected figure and table locations

Main text

• Figure 1 · p. 1
• Figure 2 · p. 3
• Table 1 · p. 6
• Figure 3 · p. 6
• Table 2 · p. 7
• Table 3 · p. 7
• Table 4 · 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 5 physical pages; 3 detected figures and 1 tables. These may include overlapping main-text material.

Detected heading	PDF page	Estimated span
1. SIFT, w/ KC, 5-NN (standard TESO, see Methods section of the paper) with	1	≈ 0.08 pp.
2. SuperGlue [6] matches (x, y) ∈ SG, w/o KC, i. e.:	1	≈ 0.07 pp.
3. SuperGlue matches (x, y) ∈ SG, w/ KC, i. e.:	1	≈ 3.89 pp.
References	5	≈ 0.6 pp.

• Figure 1 · p. 2
• Figure 2 · p. 2
• Table 1 · p. 4
• Figure 3 · p. 5

Argument and organization

1. 3. Methods

   Defines manifold parameterization, robust epipolar error, and streaming updates.

2. 4. Datasets and 5. Experiments

   Separates synthetic drift, real calibration assessment, and a nonsequential optimization test.

Problem, claim and contribution

The motivating plot links pose drift to stereo depth errors, giving calibration accuracy an observable downstream consequence.

Figure 1 p. 1 · Main paper · Additional source 2

Argument and section order

A method chain proceeds from camera-coordinate conventions and observability to loss and adaptive updates before dataset construction.

§3 pp. 3–4 · Main paper · Additional source 2

Figures, captions and visual explanation

The first figure couples a temporal tracking curve with depth maps; later evidence uses paired geometric and stereo table blocks rather than only a single scalar.

Figure 1; Tables 2–3 p. 7 · Main paper · Additional source 2

Tables, comparisons and evidence

The KITTI discussion retains an apparent contradiction: some rectification metrics improve while depth consistency worsens, motivating the intrinsic-calibration explanation.

Table 2 and §5.2 pp. 6–7 · Main paper · Additional source 2

Appendix and reproducibility

The supplement compares detector/loss choices and analyzes latency, then stratifies full TruckScenes results by environmental conditions.

Supplement pp. 1–5 · Main paper · Additional source 2

Clarity and writing conventions

Negative improvement metrics and geometric units are stated near the tables, essential because a negative number can mean better performance.

Tables 2–3 p. 7 · Main paper · Additional source 2

What to reuse—and what not to infer

Transferable lesson: Report downstream consequences alongside parameter error and explain disagreement between metrics rather than hiding it.

Boundary: The essential-matrix parameterization has observability limits; the discrete-pair experiment is not a test of the online sequence optimizer.

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

Manuscript-wide writing scan of main PDF pp. 1–8: framing, section structure, method exposition, figure/table captions, evaluation organization and conclusions; not an exhaustive technical proof audit or reproduction.

Appendix / supplement

Scanned five supplementary pages covering matching/loss alternatives, latency, per-condition results, and kernel-correlation plots.

Visually inspected PDF pages

4, 7

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 CVPR 2026 CVF proceedings PDF; retrieved 2026-09-26 · 9 pages
• Official CVPR 2026 separate supplementary PDF; retrieved 2026-09-26 · 5 pages