OpenFace – a state-of-the art tool intended for facial landmark detection, head pose estimation, facial action unit recognition, and eye-gaze estimation.
May I ask if the feature extraction from the windws GUI OpenFaceOffline.exe produced AU outputs (AU_r & AU_c) that are normalized?
I understand that under the API calls that there is a paragraph that writes "To extract AUs in a video look at FeatureExtraction.cpp for a detailed example that performs offline AU normalization, making it more accurate. This requires writing the AU predictions to file or memory so that they could later be corrected." However, i'm unsure if this automatically works for the offline program of OpenFace and if the output is already normalized.
If not, may i know if there is a way to do that for the output?
May I ask if the feature extraction from the windws GUI OpenFaceOffline.exe produced AU outputs (AU_r & AU_c) that are normalized?
I understand that under the API calls that there is a paragraph that writes "To extract AUs in a video look at FeatureExtraction.cpp for a detailed example that performs offline AU normalization, making it more accurate. This requires writing the AU predictions to file or memory so that they could later be corrected." However, i'm unsure if this automatically works for the offline program of OpenFace and if the output is already normalized.
If not, may i know if there is a way to do that for the output?
I do hope that someone can help me with this.
Thank you!