From c4532c342389baa6c19b9d9a26f3647a0aeb4a9d Mon Sep 17 00:00:00 2001 From: Vasil Zlatanov Date: Tue, 20 Nov 2018 14:52:59 +0000 Subject: Make abstract short and good --- report/metadata.yaml | 13 +++---------- 1 file changed, 3 insertions(+), 10 deletions(-) diff --git a/report/metadata.yaml b/report/metadata.yaml index c7ede78..b60aa7e 100755 --- a/report/metadata.yaml +++ b/report/metadata.yaml @@ -10,17 +10,10 @@ lang: en babel-lang: english abstract: | In this coursework we will analyze the benefits of different face recognition methods. - On one hand we will analyze PCA, Principal Components Analysis. This method - allows dimensionality reduction, obtaining a generative subspace which is very reliable for - face reconstruction. + We analyze dimensionality reduction with PCA, obtaining a generative subspace which is very reliable for face reconstruction. Furthermore, we evaluate LDA, which is able to perform reliable classification, generating a discriminative subspace, where separation of classes is easier to identify. - On the other hand LDA, Linear Discriminant Analysis, allows to perform a very reliable classification, - generating a discriminative subspace, in which the separation between classes is easier to recognize. - - In the final part we will analyze the benefits of using a combined version of the two methods using Fisherfaces. - As we will see, the PCA-LDA ensemble will obtain much more accurate results with a very high speed of computation. - - The data used includes 52 classes with 10 samples each. The number of features is 2576(since the size of the pictures is 46x56). + In the final part we analyze the benefits of using a combined version of the two methods using Fisherfaces and evaluate the benefits of ensemble learning with regards to data and feature space ranodmisation. We find that combined PCA-LDA obtains lower classification error PCA or LDA individually, while also maintaining a low computational costs, allowing us to take advantage of ensemble learning. + The dataset used includes 52 classes with 10 samples each. The number of features is 2576 (46x56). ... -- cgit v1.2.3-54-g00ecf