What do open-access papers that map face-tracker blendshapes to FACS action units report about methods and limits?

A CC BY Sensors 2026 study (PMC12899966) maps ARKit blend shapes to action units using a published guide (its ref. 11) and reports 68.3% seven-class posed-emotion accuracy against a 58.9% mean of three human raters, while stating that ARKit coefficients do not always map cleanly to FACS units. Curator qualifications: The paper is internally inconsistent on whether the method beat all three raters or two of three; cite only the mean comparison. Accuracy is for posed emotions on one device in good lighting; not per-AU validity. PLOS ONE rig study (S11, E5-E6) not reinspected. Text only; no face tracking, rig or classifier run. Limitations: N=31 convenience sample. Evidence status: statements located in the cited sources by a correlated desk review (Claude); not executed; not independently reviewed.

{
  "citations": [
    {
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC12899966/",
      "note": "PMC12899966 (Sensors 2026) abstract, s.2.3, results, s.4.4; PMC12899966 text offsets ~30016 and ~45241"
    },
    {
      "url": "https://journals.plos.org/plosone/article/file?id=10.1371%2Fjournal.pone.0231266&type=printable",
      "note": "PMC12899966 (Sensors 2026) abstract, s.2.3, results, s.4.4; PMC12899966 text offsets ~30016 and ~45241"
    },
    {
      "url": "https://www.mdpi.com/1424-8220/26/3/1060",
      "note": "PMC12899966 (Sensors 2026) abstract, s.2.3, results, s.4.4; PMC12899966 text offsets ~30016 and ~45241"
    }
  ]
}