# What is the procedure to obtain a 3D face dataset? Can anyone suggest a good 3D face dataset?

Project Description:

Input:

Frontal Face Image
Expression
Angle
Size


Details:

I have to convert the input frontal face image into 3D face and simulate the given expression (smile, cry, neutral, ..); rotate by the given angle and re-size to the given size. Again convert back into 2D face after the changes.

After googling, I found that Morphable models is a good algorithm to start with; I want a 3D face database to start & complete the implementation.

I found some 3D databases list on www.face-rec.org. Can anyone help me with buying (pricing, procedure ....) the database, and how to choose the database?

Thank you in advance..

-

## 2 Answers

I don't personally know of a good one to vouch for, but on face-rec.org/databases/ at least, isn't the "GavabDB: face database" free?

has 3 (apparently free) downloads: Brief description: GavabDB is a 3D face database. It contains 549 three-dimensional images of facial surfaces. These meshes correspond to 61 different individuals (45 male and 16 female) having 9 images for each person. The total of the individuals are Caucasian and their age is between 18 and 40 years old. Each image is given by a mesh of connected 3D points of the facial surface without texture. The database provides systematic variations with respect to the pose and the facial expression. In particular, the 9 images corresponding to each individual are: 2 frontal views with neutral expression, 2 x-rotated views (±30º, looking up and looking down respectively) with neutral expression, 2 y-rotated views (±90º, left and right profiles respectively) with neutral expression and 3 frontal gesture images (laugh, smile and a random gesture chosen by the user, respectively).

Note: Those publications that use this signature date must reference the following work: A.B. Moreno y A.Sanchez. GavabDB: A 3D Face Database. Proc. 2nd COST Workshop on Biometrics on the Internet: Fundamentals, Advances and Applications, C. Garcia et al (eds): Proc. 2nd COST Workshop on Biometrics on the Internet: Fundamentals, Advances and Applications, Ed. Univ. Vigo, pp. 77-82, 2004

Readme.rtf (9 kB; .rtf)

CD1_faces1-20.zip (38 MB; .zip)

CD2_faces21-40.zip (39 MB; .zip)

CD3_faces41-61.zip (38 MB; .zip)

-

BTW, in case you are interested in applying face algos to general objects, Berkeley has 3-D Object Dataset: http://kinectdata.com/

-
Thank You for the reply, ....I have downloaded the dataset. Only shape information (the vertices of the mesh) are given, texture information is not shared. Is texture information must for 3D face reconstrction?? If possible can you refer any paper/algorithm where texture information of the 3D datasets while modeling is not necessary? –  user1317084 Dec 31 '12 at 4:30