Two frequent problems in content-based image retrieval systems are the poor quality of the results, due to the semantic gap, and the slowness of the image retrieval. For controlling these problems when human face image sets are considered, a new approach, called the MIFLIR system, is presented in this paper. This approach allows reducing the semantic gap by… (More)
Face recognition is typically an ill-posed problem because of the limited number of available samples. As experimental results show, combining multiclassifier fusion with the RBPCA MaxLike approach, which couples covariance matrix regularization and block-based principal component analysis (BPCA), can provide an effective framework for face recognition that… (More)
Many researches have used convolutional neural networks for face classification tasks. Aiming to reduce the number of training samples as well training time, we propose to use a LSTM network and compare its performance with a standard MLP network. Experiments with face images from CBCL database using PCA for feature extraction provided good results… (More)
In this technical poster, we present the first results of our research on face feature detection. The main goal is to build a tool capable of producing a set of suitable data for face similarity comparisons on content based retrieval systems. The results are still not very precise, but quite promising.