Kinship Recognition

UB KinFace Database

Version 2

Last modified: July 31, 2011

Database Description

UB KinFace database is used to develop, test, and evaluate kinship verification and recognition algorithms. It comprises 600 images of 400 people which can be separated into 200 groups. Each group is composed of child, young parent and old parent images. Most of images in the database are real-world collections of public figures (celebrities and politicians) from Internet. To the best of our knowledge, it is the first database that contains all children, young parents and old parents for the purpose of kinship verification.

Figure 1. UB KinFace database illustration on the distributions of four kinship relations over two different race categories.

UB KinFace database can be mainly divided into two parts in terms of race, Asian and Non-Asian, each of which consists of 100 groups, 200 people and 300 images. Typically, there are four kinship relations, namely, “son-father”, “son-mother”, “daughter-father” and “daughter-mother”, as shown in Figure 1. In addition, Figure 2 illustrates the statistics from the perspective of race and kinship relations.

Figure 2. Statistical information of the UB KinFace database.

Terms of Use

This dataset is for non-commercial research purposes only. The image copyright belongs to the original author or the media as listed in the following URL file. If you find this collection useful for your research, please cite the papers below.

Citation

1. Siyu Xia, Ming Shao, Jiebo Luo, and Yun Fu, “Understanding Kin Relationships in a Photo”, IEEE Transactions on Multimedia (T-MM), accepted, 2012.

2. Siyu Xia, Ming Shao and Yun Fu, “Kinship Verification through Transfer Learning,” International Joint Conferences on Artificial Intelligence (IJCAI), pp. 2539-2544, 2011

3. Ming Shao, Siyu Xia and Yun Fu, “Genealogical Face Recognition based on UB KinFace Database,” IEEE CVPR Workshop on Biometrics (BIOM), 2011

Bibtex

@article{Buffalo_TMM_Kinship,

author = { S. Xia, M. Shao, J. Luo and Y. Fu },

title = {Understanding Kin Relationships in a Photo},

journal = {IEEE Transactions on Multimedia},

volume = { },

number = { },

year = {2012},

pages = { -- },

}

@inproceedings{Siyu_IJCAI11_Kinship,

author = {S. Xia, M. Shao and Y. Fu},

title = {Kinship Verification through Transfer Learning},

booktitle = {Proc. International Joint Conferences on Artificial Intelligence (IJCAI)},

pages={2539--2544},

year = {2011},

}

@inproceedings{Ming_CVPR11_Genealogical,

author = {M. Shao, S. Xia and Y. Fu},

title = {Genealogical Face Recognition based on UB KinFace Database},

booktitle = {Proc. IEEE CVPR Workshop on Biometrics (BIOM)},

year = {2011},

}

Data Files

o A database file: [KinFace_V2]

o An image link file containing the original image URL: [KinFace_V2_URL]

o A textual database with coordinates of eye centers, nose tip and mouth center: [KinFace_V2_Coordinate]

SCface_coord1

Figure 3. (Left) Coordinates of the eye centers, nose tip and mouth. (Right) Their corresponding textual database entries.

Online Annotation Tool

We developed an online annotation tool to ask voluntary Web visitors to help collect and tag the kinship data. The video demo http://youtu.be/LDJihDhN-qc shows our motivation, the interface of the online annotation tool, and the usage of tagging the relation of people in an image. The tool will be available soon.