Table 10 EER results for non-twins using standard FERET subsets under expression, illumination and age variations
From: Recognition of identical twins using fusion of various facial feature extractors
EER results (%) of | |||||
|---|---|---|---|---|---|
Challenge | Expression | Illumination | Aging | Aging | |
subset | (fb) | (fc) | (duplicate I) | (duplicate II) | |
Algorithms | CNN | 6.10 | 26.44 | 9.88 | 13.54 |
PCA | 11.4 | 34.3 | 23.2 | 30.7 | |
HOG | 9.5 | 31.9 | 13 | 14 | |
LBP | 3.43 | 24.9 | 11.15 | 12 | |
Score fusion (PCA, HOG, LBP) | 6.2 | 22.7 | 12.6 | 14.67 | |
Score fusion (HOG, LBP) | 5.9 | 24.2 | 10.25 | 10.7 | |
Feature fusion (PCA, HOG, LBP) | 8.7 | 28.1 | 11.15 | 14 | |
Feature fusion (HOG, LBP) | 3.9 | 24.5 | 10.75 | 12 | |
Decision fusion (PCA, HOG, LBP) | 2.9 | 21.8 | 8.15 | 8.7 | |
Proposed method | 2.5 | 20.6 | 7.5 | 8 | |