Writing study: FEEL (Force-Enhanced Egocentric Learning): A Dataset for Physical Action Understanding
← ECCV 2026 oral writing corpus
Eadom Dessalene, Botao He, Michael Maynord, Yonatan Tussa, Pavan Mantripragada, Yianni Karabatis, Nirupam Roy, Yiannis Aloimonos · Official ECCV 2026 oral record
Figure, table and section ledger
Editorial contribution type: Datasets, benchmarks & measurement (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
- 15 physical pages
- Before references / appendix
- ≈ 12.51 page-equivalents
- Main figures / tables
- 7 / 4
- Attached appendix figures / tables
- 0 / 0
- Attached appendix span
- No boundary detected; not assumed zero
Extraction notes: No attached appendix boundary detected. Zero means no recognized caption, not independently verified absence.
Source-bound correction: All top-level main headings and the reference boundary checked against source PDF text; source-bound positions use physical single-column y/page-height, not a two-column reading-order surrogate. Running page-number/author titles, affiliations, numbered contribution bullets, footnotes and bibliography continuations are excluded. This linked manuscript contains no appended supplementary section; separate supplement links were not scanned. Selected disputed heading locations also checked in rendered source pages. Tables 1–3 are real top-of-page caption labels (pp.7–9), retained despite the unusually small page top margin in this source; source text and bounding boxes checked.
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.
All detected figure and table locations
Main text
Attached appendix
No recognized caption labels.
Argument and organization
- 1 Introduction; 2 Related Work
Uses force as a physical signal for contact rather than only visual proximity, then connects that distinction to more precise supervision.
- 3 Methods: dataset collection, contact understanding and forces from video
Introduces the capture hardware and synchronization before label construction, learning tasks and transfer benchmarks, preserving the provenance of supervision.
- 4 Experiments; 5 Discussion and limitations
The action-transfer table pairs original versus additional force pretraining within each backbone and separates frozen from unfrozen evaluation; not every entry improves.
Problem, claim and contribution
Uses force as a physical signal for contact rather than only visual proximity, then connects that distinction to more precise supervision.
Abstract and Section 1, pp.1–3 · Main paper
Argument and section order
Introduces the capture hardware and synchronization before label construction, learning tasks and transfer benchmarks, preserving the provenance of supervision.
Sections 3–4, pp.4–10 · Main paper
Figures, captions and visual explanation
A labeled glove photograph locates sensors and electronics; a separate temporal strip shows the predicted contact transition with the same event repeated over time.
Figure 4, p.5; Figure 6, p.10 · Main paper
Tables, comparisons and evidence
The action-transfer table pairs original versus additional force pretraining within each backbone and separates frozen from unfrozen evaluation; not every entry improves.
Table 4, p.10 · Main paper
Appendix and reproducibility
The main manuscript points to supplementary synchronization calibration, thresholds and pseudo-label details, without making those details a reviewed separate source here.
Sections 3.1–3.2, pp.6–8 · Main paper
Clarity and writing conventions
Dataset comparisons specify modality codes and partial annotations; the limitations paragraph names mechanical drift, comfort and stationary-object constraints.
Table 1, p.7; Limitations, p.13 · Main paper
What to reuse—and what not to infer
Transferable lesson: A sensor dataset paper needs hardware and synchronization evidence, label-ambiguity handling, a dataset comparison and downstream tests that isolate added supervision.
Boundary: Force-derived labels and motion-based segmentation pseudo-labels have different supervision assumptions; improved average transfer does not imply gains in every task.
Apply this lesson using the argument and evidence worksheet.
Inspection record and sources
- Checked
- 2026-10-02
- Reading status
- Writing-focused manuscript scan + selected visuals inspected
- Main text
- Writing-oriented scan across the manuscript main argument, physical PDF pp.1–13: framing and section roadmaps, method/argument explanations, figure and table captions, comparison protocols, conclusions and limitations. Selected full-page rasters were inspected. This is not exhaustive line-by-line technical reading, proof verification or experimental reproduction.
- Appendix / supplement
- Main-manuscript pointers to appendix/supplement and the reference boundary were checked. Separate supplementary PDFs were not retrieved or inspected; their contents are not included in inspected-page counts.
- Visually inspected PDF pages
- 5, 10
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.