From b54ec48d96328bcf327f84272a8b80aabe1afc11 Mon Sep 17 00:00:00 2001 From: nunzip Date: Wed, 7 Nov 2018 17:44:38 +0000 Subject: Add peak accuracy value --- report/paper.md | 18 +++++++++++++++--- 1 file changed, 15 insertions(+), 3 deletions(-) (limited to 'report/paper.md') diff --git a/report/paper.md b/report/paper.md index 113dfa6..e201252 100755 --- a/report/paper.md +++ b/report/paper.md @@ -38,8 +38,8 @@ for our standard seed can be observed below. To perform face recognition we choose the best M eigenvectors associated with the largest eigenvalues. We tried different values of M, and we found an optimal point for -M=120. After such value the accuracy starts to flaten, with -some exception for points at which accuracy decreases. +M=42 with accuracy=66.3%. After such value the accuracy starts +to flaten, with some exceptions for points at which accuracy decreases. WE NEED TO ADD PHYSICAL MEANINGS ![Recognition Accuracy of Test data varying M](fig/accuracy.pdf "Accuracy1") @@ -80,7 +80,19 @@ PCA &Fast PCA\\ It can be proven that the eigenvalues and eigenvectors obtain are the same: ##PROVE -Reconstruction is then performed on a chosen +Using the computational method for fast PCA, face reconstruction is then performed. +The quality of reconstruction will depend on the amount of eigenvectors picked. +The results of varying M can be observed in the picture below. A face from class +number 21 is reconstructed as shown below withrespective M values of M=10, M=100, +M=200, M=300. The last picture is the original face. + +![Reconstructed Face](fig/face160rec.pdf) + +It is already observable that the improvement in reconstruction is marginal for M=200 +and M=300. For such reason choosing M close to 100 is good enough for such purpose. + +IT HAS TO BE DONE FOR MORE FACE IMAGES + # Cites -- cgit v1.2.3-54-g00ecf