Transformer-based Human Action Recognition using Skeleton Heatmap
Bibliographic record
Abstract
Human activity recognition (HAR) can alert clinicians when a frail patient falls, track rehabilitation exercises, and ease the workload of healthcare staff. Yet continuous video recording in wards or homes raises serious privacy concerns. Converting each frame to an anonymous 3D skeleton keeps only the motion cues needed for HAR while removing faces, clothing, and scene details. We extend SwinPoseFormer, a transformer that operates on skeleton heatmaps beyond its original FineGym benchmark to two larger and more varied datasets: Kinetics-400 and NTU-RGB+D 120. Using the same lightweight pipeline (incorporating uniform temporal sampling, subject-centred cropping, Gaussian joint and limb heatmaps generation, and a single-channel input strategy), the model achieves 78.3% / 98.7% top-1 / top-5 accuracy on Kinetics-400 and 74.0% / 98.0% on NTU-120, compared to 84.3% / 98.5% on FineGym. The stable top-5 scores demonstrate strong cross-dataset generalization without leaking visual identity. These results show that privacy-preserving skeleton streams captured by inexpensive IoT cameras can support reliable, real-time HAR across a broad range of daily activities.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.003 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".