Table 7 Local processing based 3D face recognition approaches [6]
Category/references | Dataset | Local sub-regions | Identification rate (%) | Verification rate (%) | Classifier/feature detection |
|---|---|---|---|---|---|
Chang et al. [4] | ND2006 | Interior nose, center face,entire nose | 87.10 | – | PCA |
Kin-Chung et al. [81] | FRGC v 2.0 | 10 Sub-regions from nose, eyes, cheeks and eyebrows | – | 90.45 | LDA,LSVM |
Faltemier et al. [83] | FRGC v 2.0 | 38 Sub-regions from nose, eyes, forehead and chin | 97.20 | 93.2 | – |
Boehnen et al. [91] | FRGC v 2.0 | Eight sub-regions around eyes and nose | 95.50 | – | Nearest neighbor |
Queirolo et al. [75] | FRGC v 2.0 | Sub-regions from nose and forehead | 98.40 | 96.50 | – |
Spreeuwers et al. [92] | FRGC v 2.0 | 30 Overlapping sub-regions from all over the face | 99.00 | 94.60 | PCA-LDA |
Li and Da [93] | FRGC v 2.0 | Six sub-regions: forehead, left mouth, right mouth, nose,left cheek and right cheek | 97.80 | 96.00 | PCA |
Lei et al. [94] | FRGC v 2.0 | Sub-regions from nose, eyes and forehead area | 95.60 | 97.60 | Modified LDA, polynomial-linear-RBF SVM |
Tang et al. [87] | BU-3DFE | Special feature based sparse division (29 blocks) and dense division (59 blocks) | 97.70 | 98.20 | Nearest neighbor |
FRGC v2.0 | 94.89 | – | |||
Li et al. [95] | BJUT-3D | Multi-scale multi resolution patches (MSMC-LNP) | 95.10 | – | W-SRC |
FRGC v2.0 | 96.30 | – | |||
BU-3DFE | 92.21 | – | |||
Bosphorus | 95.40 | – | |||
Ming [89] | FRGC v2.0 | Seven sub-regions: forehead, mouth, left cheek, right cheek, nose, left eye and right eye regions | – | 95.03 | – |