EUVIMEDEUVIMED entdecken
Crossref

Calibrated VideoLevel Scoring of Skeleton Reconstruction Residuals for Campus Safety Monitoring

Zheqi Lyu, Ke Chen

Humanities and Social Sciences · 2026 · 2 (4)

DOI: 10.65285/hss.v2i4.005 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Humanities and Social Sciences, 2026. Verfasst von Zheqi Lyu und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Skeletonbased anomaly detection provides a lightweight approach to humanmotion analysis in campus safety monitoring. Campus safety events are often brief and temporally sparse, posing a challenge for conventional temporal averaging. In this paper, we propose a videolevel residual scoring protocol for skeletonbased anomaly detection. Without modifying the underlying reconstruction architecture, the protocol calibrates local residual responses using normalvalidation statistics and aggregates highresponse temporal segments through uppertail selection. Under a skeletonobservable safetyevent evaluation setting on ShanghaiTech, our method improves the ROCAUC of LSTMAE from 0.789 to 0.840 and that of GRUAE from 0.783 to 0.825 when the uppertail ratio is selected automatically using only normal validation data. Ablation studies show that uppertail aggregation provides the main performance improvement by retaining high-response temporal segments, while regional calibration further improves residual comparability. Supp

Zitieren

APA

Lyu, Z., & Chen, K. (2026). Calibrated VideoLevel Scoring of Skeleton Reconstruction Residuals for Campus Safety Monitoring. Humanities and Social Sciences, 2(4). https://doi.org/10.65285/hss.v2i4.005

Vancouver

Lyu Z, Chen K. Calibrated VideoLevel Scoring of Skeleton Reconstruction Residuals for Campus Safety Monitoring. Humanities and Social Sciences. 2026;2(4). doi:10.65285/hss.v2i4.005

BibTeX
@article{Lyu2026,
  author = {Lyu, Zheqi and Chen, Ke},
  title = {Calibrated VideoLevel Scoring of Skeleton Reconstruction Residuals for Campus Safety Monitoring},
  journal = {Humanities and Social Sciences},
  year = {2026},
  volume = {2},
  number = {4},
  doi = {10.65285/hss.v2i4.005}
}
Kostenlos starten →