Menu

It has 1564 groups of pictures in total a 14

It has 1564 groups of pictures in total a 14,126 that consists of 1199 persons and 365 duplicate groups of pictures. A duplicate set is another group of pictures of an individual present in the database that was generally captured on a dissimilar day. This database can be downloaded from the link http://www.nist.gov/humanid/feret/. The color FERET dataset can be downloaded from the link http://www.nist.gov/humanid/colorferet/.

Fig. 5 pose variations images from FERET database 6.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

G. Korean Face Database (KFDB)

It consists of facial pictures of lots of Korean individuals gathered below constrained environments. In this, pictures with changing pose, lighting, and facial expressions were captured. The individuals were photographed within the middle of an octagonal structure and also the cameras were set between 450 off frontal in 2 orientations at 150 increments.

Fig. 6 Variations in pose from the Korean face database 2.

H. Yale Face Database B

It was the information gathered for the economical testing of face recognition techniques below appreciable changes in lighting and pose. The individuals were pictured within a geodesic dome by sixty four computer-controlled xenon strobes. The pictures of ten individuals were captured below sixty four lighting conditions in nine different poses. This database can be downloaded from the link http://cvc.yale.edu/projects/yalefacesB/yalefacesB.html.

Fig. 7 Variation in illuminations from Yale Face Database B 6.

I. Yale Face Database

It consists of eleven pictures of fifteen people in a different situations having with and without glasses, varies in expression of faces and lighting variation 2. This database can be downloaded from the link http://cvc.yale.edu/projects/yalefaces/yalefaces.html.

J. CMU Pose, Illumination, and Expression (PIE) Database

This database efficiently samples a several pose and lighting situations and different facial expressions. It has made an influence on development of algorithm for recognition of faces across pose. It consists of 41,368 pictures captured from 68 person. The RGB color pictures are 640 × 480 in size 3. This database can be downloaded from the link http://www.ri.cmu.edu/projects/project 418.html.

Fig. 8 Illumination variation images from PIE face database 3.

K. SCface Database

This database consists of stationary pictures of human faces. The pictures were captured in unconstrained conditions with the help of 5 video surveillance cameras of different features. It has 4160 stable pictures of one hundred thirty people. The pictures from several quality cameras create the practical-world situations which helps in robust face recognition algorithms testing. It is freely available to research community 3.

L. Georgia Tech Face Database

It consists of pictures of fifty person which are kept in JPEG format. In this, most of the pictures were captured in 2 dissimilar time spans to think about the changes in lighting conditions, facial expression, and look. Also, the faces were taken at dissimilar scales and directions. Every picture is manually marked to find the location of the face in the picture.

M. Japanese Female Facial Expression (JAFFE) Database

It consists of two hundred thirteen pictures of seven facial expressions (6 normal + 1 neutral) in various poses by ten Japanese feminine models. In this, all photos has been evaluated on six feeling options by sixty Japanese individuals 2. It can be downloaded from the link http://www.mis.atr.co.jp/˜mlyons/jaffe.html.

Fig. 9 Expression variation images from JAFFE database 2.

N. Indian Face Database

It has eleven different pictures of fourty different persons. All the pictures are stored in JPEG format. The size of every picture is 640 x 480 pixels, which are having 256 grey scales for each pixel. The pictures are arranged in 2 broad categories – males and females. The various directions of the face included are: seeing front, left, right, up, up towards left, up towards right, down. The different emotions present are: smile, neutral, sad/disgust, laughter 3. It can be downloaded from the link http://www.cs.umass.edu/~vidit/facedatabase.

Fig. 10 Pose variation images from Indian Face database 3.

O. FEI Face Database

It is a face database which consists of a collection of face pictures captured at FEI in Brazil. It has fourteen pictures of all the two hundred persons, which gives a total of two thousand eight hundred pictures.

P. The Bosphorus database
It is a completely unique 3D face database that encompasses a giant set of expressions, coherent changes of poses and various kinds of occlusions. It is very useful for the advancement and analysis of approaches on recognition of faces under unfavorable environments, facial expression evaluation and synthesis.
Q. FaceScrub Database

The information of the database gathered from the pictures present on the web. It has an automatic procedure which justifies that the picture associates to the correct person. It has the pictures of five hundred thirty persons which is a total of 107,818. The pictures are supplied with the name and gender notations.

III. RECENT FACIAL DATABASES

The earlier databases were meant on facial detection for people recognition, the latest databases are tuned more towards collecting the changes in picturing methods, expressions of faces, and anonymities because of makeup. Few of the recent facial databases are mentioned in 7 are:

A. Labelled Wikipedia Faces (LWF)

It consists of mined pictures from more than 0.5 million biographic records from the Wikipedia Living folks records. It consists of 8500 faces from 1500 persons. YouTube Faces Database (YFD) consists of 3425 videos of 1595 different persons (2.15 videos for each person) with video clips starting from forty eight to 6070 frames. It was designed to convey a cluster of videos and marks for person’s recognition from videos.

YouTube Makeup Dataset (YMD)

It contains an information that accommodates pictures from 151 persons from YouTube makeup tutorials previous and once precise to substantial makeup. During this process, four shots were captured for all persons (two shots previous and 2 shots once makeup). It has constant illumination however it illustrates the difficulties in recognitions of face before changes in makeup.

B. Indian Movie Face Database (IMFD)

It contains an information that has 34512 pictures from a hundred Indian actors collected from regarding a hundred videos and cropped to own variations in illumination, pose, expression, resolution, occlusions and makeup.

IV. COMPARISON OF FACE DATABASES

The several face databases have been constructed for the evaluation of faces when associating with one or an integration of these deviations. The several kinds of face databases are mentioned in the Table1.

Table 1. Several kinds of face databases 5, 6

Face Database Picture Type RGB/Gray Scale
Picture Size Kinds of
constraints

FERET
Gray RGB
256 x 384
I/O, i, t, e, p

The yale face B
Gray Scale
640 x 480
p, i

AR Faces
RGB
576 x 768
t, e ,i , o

CMU-PIE
RGB
640 x 486
P, i, e
The yale face
Gray Scale
320 x 243
e, i

Asian face database
RGB
640 x 480
i, o, p, e

Indian face database
RGB
640 x 480
P, e
The several picture deviations are given as, p: pose, i:
illumination, o: occlusion, e: expression, t: time delay, I/O: indoor/outdoor conditions.

V. CONCLUSION
Face recognition in unconstrained situations continues to be a challenging research domain. This paper have presented an extensive survey of face databases for constrained and unconstrained environments. It focuses mostly on novel databases that are freely available for the research purposes. Most of the popular face databases are concisely introduced and compared. The purpose of this review paper is to assist the young budding researchers in the area of face recognition by compiling the most widely used face databases and the link to download them so as to motivate their further research.

x

Hi!
I'm Sam!

Would you like to get a custom essay? How about receiving a customized one?

Check it out