yaohungt / Multimodal-Transformer

[ACL'19] [PyTorch] Multimodal Transformer
MIT License
800 stars 150 forks source link

About 35 Facial Action Units #29

Closed xingbow closed 3 years ago

xingbow commented 3 years ago

Hi, thanks for your processed data. I want to ask what 35 facial action units correspond to since there are more than 35 facial action units in the FACET iMotions. https://imotions.com/blog/facial-action-coding-system/

xingbow commented 3 years ago

I found the metadata of the processed data from the CMU-Multimodal SDK: https://github.com/A2Zadeh/CMU-MultimodalSDK

"dimension names": [
    "Anger",
    "Contempt",
    "Disgust",
    "Joy",
    "Fear",
    "Baseline",
    "Sadness",
    "Surprise",
    "Confusion",
    "Frustration",
    "AU1",
    "AU2",
    "AU4",
    "AU5",
    "AU6",
    "AU7",
    "AU9",
    "AU10",
    "AU12",
    "AU14",
    "AU15",
    "AU17",
    "AU18",
    "AU20",
    "AU23",
    "AU24",
    "AU25",
    "AU26",
    "AU28",
    "AU43",
    "Has_Glasses",
    "Is_Male",
    "Pitch",
    "Yaw",
    "Roll"
  ],