Writing study: SDTrack: A Baseline for Event-based Tracking via Spiking Neural Networks

CVPR overviewCorpus atlasBy contribution typeVisual galleryRules & templateWriting patternsPaper studies

← CVPR 2026 oral writing corpus

Yimeng Shan, Zhenbang Ren, Haodi Wu, Wenjie Wei, Rui-Jie Zhu, Shuai Wang, Dehao Zhang, Yichen Xiao, Jieyuan Zhang, Kexin Shi, Jingzhinan Wang, Jason K. Eshraghian, Haicheng Qu, Malu 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: Efficiency, inference & 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

10 physical pages

Before references / appendix

≈ 8.11 page-equivalents

Main figures / tables

8 / 4

Attached appendix figures / tables

0 / 0

Attached appendix span

No boundary detected; not assumed zero

Before references / appendix: ≈ 81.1%References / intervening material: ≈ 18.9%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	≈ 1.25 pp.	12.5%
2. Related Work	2	≈ 0.64 pp.	6.4%
3. SDTrack Pipeline	3	≈ 2.39 pp.	23.9%
4. Experiments	5	≈ 2.21 pp.	22.1%
5. Discussion and Limitation	7	≈ 0.94 pp.	9.4%
6. Conclusion	8	≈ 0.21 pp.	2.1%
7. Acknowledgments	9	≈ 0.06 pp.	0.6%
References	9	≈ 1.89 pp.	18.9%

All detected figure and table locations

Main text

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

Attached appendix

No recognized caption labels.

Argument and organization

1. 3. SDTrack Pipeline

   Introduces SNN preliminaries, event aggregation, and the tracker.

2. 4–5. Experiments and Discussion

   Separates standard tracking comparisons, component tests, alternative architectures, and deployment limitations.

Problem, claim and contribution

The first comparison combines tracking quality, parameter count, and energy, making the desired trade-off explicit from the outset.

Figure 1 p. 1 · Main paper

Argument and section order

A full pipeline with module insets precedes a concrete event-aggregation example and equations.

Figure 2 p. 3; §3.2 · Main paper

Figures, captions and visual explanation

The aggregation figure uses a direction-changing event stream to explain what different encodings preserve; AUC/precision curves and sensitivity sweeps answer separate evaluation questions.

Figure 3 p. 4; Figures 5–6 p. 7 · Main paper

Tables, comparisons and evidence

Ablations replace positional encodings and tracking-head choices under aligned settings; the comparison states when competing trackers also use the proposed aggregation.

Table 2 and §4.3–4.4 p. 7 · Main paper

Appendix and reproducibility

The text refers to additional dataset specifications, but no separate supplement was available for inspection.

§4.3 p. 7 · Main paper

Clarity and writing conventions

The discussion explicitly reserves actual neuromorphic-chip deployment for future work, qualifying the theoretical energy-efficiency story.

§5 p. 8 · Main paper

What to reuse—and what not to infer

Transferable lesson: For neuromorphic efficiency, show what information the representation retains and keep modeled energy distinct from deployment measurements.

Boundary: Operation-based theoretical energy estimates are not measured system power; the manuscript itself notes future chip deployment.

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

No separate supplementary PDF was available at this scan; referenced extra benchmark details were not independently inspected.

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 · 10 pages