This is a project of facial(lips) experssinion recognition. Features: four kinds of classification: none, openmouth, pouting, smile Model: 3 conv 3 full-connected layers, train_acc:92.82%, valid_acc:94.30%. Required Environment: Dlib, Pytorch, Anaconda3 You can test a image in test.py, or test a video in video_facialexpression_predict.py, or test realtime video from own camera in Mycamrea_facialexperssion_predict.py I'm sorry to tell you that 'shape_predictor_68_face_landmarks.dat' is not avaliable due to upload file size limitation.
-
Notifications
You must be signed in to change notification settings - Fork 0
. This is a project of facial(lips) experssinion recognition. Features: four kinds of classification: none, openmouth, pouting, smile Model: 3 conv 3 full-connected layers, train_acc:92.82%, valid_acc:94.30%. Required Environment: Dlib, Pytorch, Anaconda3
zzg-tju/Facial-lips-Expression-Recognition
About
. This is a project of facial(lips) experssinion recognition. Features: four kinds of classification: none, openmouth, pouting, smile Model: 3 conv 3 full-connected layers, train_acc:92.82%, valid_acc:94.30%. Required Environment: Dlib, Pytorch, Anaconda3
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published