Classroom incident recognition is substantially harder than established general action-recognition benchmarks, with PoseC3D dropping from over 94% accuracy to 70.54%¶
Claim · All claims
Evidence · 1 study · designr2·q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i? PoseC3D, reported to achieve over 94% accuracy on NTURGBD and FineGym, reaches only 70.54% on the classroom incident dataset, which the authors attribute to viewpoint variation, subject diversity, noisy pose trajectories, and fine-grained differences between safety-related actions. → Paritosh Parmar 2026
Evidence¶
Paritosh Parmar 2026¶
Paritosh Parmar, Landy Lan, Hong Yang, Chen Yi, Chiat Pin Tay. (2026). Robust and Efficient Motion Reasoning for Privacy-Aware Classroom Incident Recognition. arXiv preprint. https://arxiv.org/abs/2608.05115 doi:10.48550/arxiv.2608.05115
q2 · i? · design · r2
Cross-benchmark comparison in the synthetic-dataset results section: the same PoseC3D model scores over 94% on established benchmarks but only 70.54% on the classroom incident dataset. The authors state the task "remains far from solved"; accuracies only, no effect size.
"PoseC3D [11] has been reported to achieve over 94% accuracy on established benchmarks such as NTURGBD [26] and FineGym [37], but reaches only 70.54% accuracy on our classroom incident dataset."
Discussion¶
Related Claims¶
- A distilled multi-order kinematic model outperforms pose-based baselines on a synthetic classroom incident dataset while using one-tenth as many parameters as PoseC3D — related
- Models trained on synthetic classroom incident data transfer zero-shot to real-world footage, with the proposed method best at 63.41% though a synthetic-to-real domain gap remains — related
- Distilling a multi-order hierarchical teacher into a smaller zeroth-order student lets the student outperform the teacher, and improves 4.1 percentage points in accuracy over STGCN++ — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
classroom-incident-recognition-harder-than-general-benchmarks |
title |
Classroom incident recognition is substantially harder than established general action-recognition benchmarks, with PoseC3D dropping from over 94% accuracy to 70.54% |
description |
Classroom incident recognition is substantially harder than established general action-recognition benchmarks, with PoseC3D dropping from over 94% accuracy to 70.54% |
status |
review |
evidence_strength |
weak |
newest_evidence |
2026 |
recheck_by |
2029 |
generated |
by process:wiki-ingest on 2026-10-11 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — paritosh-parmar-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.