-
Notifications
You must be signed in to change notification settings - Fork 0
This is a practical activity of Topics in Computer Vision course
lauraslopes/face_recognizer
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
Relatório trabalho 1 Visão Computacional Aluna: Laura Silva Lopes, GRR20163048 Os caminhos (linha 149 e linha 150) até os diretórios com as imagens estão como: path_yale = 'yale_faces' path_orl = 'orl_faces' Estão implementados e executando corretamente no código: - Leitura do banco de imagens - Computação da face média - Computação das eigenfaces e implementação do método de reconhecimento pela Eigenface - Uso de validação cruzada para o reconhecimento Os datasets usados e suas características são: ORL (AT&T) Database -- 10 pessoas x 40 imagens, cada uma com 92x112 pixels e 256 níveis de cinza The Yale Face Database -- 15 pessoas x 11 imagens, cada uma com 320x243 pixels A saída obtida foi: YALE dataset The accuracy for .centerlight images is 0.266666666667 The accuracy for .glasses images is 0.666666666667 The accuracy for .happy images is 0.8 The accuracy for .leftlight images is 0.4 The accuracy for .noglasses images is 1.0 The accuracy for .normal images is 0.866666666667 The accuracy for .rightlight images is 0.266666666667 The accuracy for .sad images is 0.866666666667 The accuracy for .sleepy images is 0.933333333333 The accuracy for .surprised images is 0.533333333333 The accuracy for .wink images is 0.666666666667 The best image to classify is .noglasses and the worst is .centerlight ORL dataset The accuracy for images 1 is 0.558823529412 The accuracy for images 2 is 0.741935483871 The accuracy for images 3 is 0.617647058824 The accuracy for images 4 is 0.592592592593 The accuracy for images 5 is 0.612903225806 The accuracy for images 6 is 0.59375 The accuracy for images 7 is 0.5 The accuracy for images 8 is 0.416666666667 The accuracy for images 9 is 0.470588235294 The accuracy for images 10 is 0.483870967742 Mean accuracy: 0.558877776021 Stand deviation accuracy: 0.0890770017773 The training and testing took 100.396633863 seconds
About
This is a practical activity of Topics in Computer Vision course
Topics
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published