Comments on the “Kinship Face in the Wild” Data Sets

Miguel Bordallo Lopez, Elhocine Boutellaa, Abdenour Hadid

Research output: Contribution to journalArticleScientificpeer-review

14 Citations (Scopus)

Abstract

The Kinship Face in the Wild data sets, recently published in TPAMI, are currently used as a benchmark for the evaluation of kinship verification algorithms. We recommend that these data sets are no longer used in kinship verification research unless there is a compelling reason that takes into account the nature of the images. We note that most of the image kinship pairs are cropped from the same photographs. Exploiting this cropping information, competitive but biased performance can be obtained using a simple scoring approach, taking only into account the nature of the image pairs rather than any features about kin information. To illustrate our motives, we provide classification results utilizing a simple scoring method based on the image similarity of both images of a kinship pair. Using simply the distance of the chrominance averages of the images in the Lab color space without any training or using any specific kin features, we achieve performance comparable to state-of-the-art methods. We provide the source code to prove the validity of our claims and ensure the repeatability of our experiments.
Original languageUndefined
Pages (from-to)2342 - 2344
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Volume38
Issue number11
DOIs
Publication statusPublished - 1 Nov 2016
MoE publication typeA1 Journal article-refereed

Keywords

  • kinship verification
  • face recognition
  • biometrics

Cite this

@article{77779cd01e314558964e1c3c802353d1,
title = "Comments on the “Kinship Face in the Wild” Data Sets",
abstract = "The Kinship Face in the Wild data sets, recently published in TPAMI, are currently used as a benchmark for the evaluation of kinship verification algorithms. We recommend that these data sets are no longer used in kinship verification research unless there is a compelling reason that takes into account the nature of the images. We note that most of the image kinship pairs are cropped from the same photographs. Exploiting this cropping information, competitive but biased performance can be obtained using a simple scoring approach, taking only into account the nature of the image pairs rather than any features about kin information. To illustrate our motives, we provide classification results utilizing a simple scoring method based on the image similarity of both images of a kinship pair. Using simply the distance of the chrominance averages of the images in the Lab color space without any training or using any specific kin features, we achieve performance comparable to state-of-the-art methods. We provide the source code to prove the validity of our claims and ensure the repeatability of our experiments.",
keywords = "kinship verification, face recognition, biometrics",
author = "Lopez, {Miguel Bordallo} and Elhocine Boutellaa and Abdenour Hadid",
year = "2016",
month = "11",
day = "1",
doi = "10.1109/TPAMI.2016.2522416",
language = "Undefined",
volume = "38",
pages = "2342 -- 2344",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "Institute of Electrical and Electronic Engineers IEEE",
number = "11",

}

Comments on the “Kinship Face in the Wild” Data Sets. / Lopez, Miguel Bordallo; Boutellaa, Elhocine; Hadid, Abdenour.

In: IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 38, No. 11, 01.11.2016, p. 2342 - 2344.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Comments on the “Kinship Face in the Wild” Data Sets

AU - Lopez, Miguel Bordallo

AU - Boutellaa, Elhocine

AU - Hadid, Abdenour

PY - 2016/11/1

Y1 - 2016/11/1

N2 - The Kinship Face in the Wild data sets, recently published in TPAMI, are currently used as a benchmark for the evaluation of kinship verification algorithms. We recommend that these data sets are no longer used in kinship verification research unless there is a compelling reason that takes into account the nature of the images. We note that most of the image kinship pairs are cropped from the same photographs. Exploiting this cropping information, competitive but biased performance can be obtained using a simple scoring approach, taking only into account the nature of the image pairs rather than any features about kin information. To illustrate our motives, we provide classification results utilizing a simple scoring method based on the image similarity of both images of a kinship pair. Using simply the distance of the chrominance averages of the images in the Lab color space without any training or using any specific kin features, we achieve performance comparable to state-of-the-art methods. We provide the source code to prove the validity of our claims and ensure the repeatability of our experiments.

AB - The Kinship Face in the Wild data sets, recently published in TPAMI, are currently used as a benchmark for the evaluation of kinship verification algorithms. We recommend that these data sets are no longer used in kinship verification research unless there is a compelling reason that takes into account the nature of the images. We note that most of the image kinship pairs are cropped from the same photographs. Exploiting this cropping information, competitive but biased performance can be obtained using a simple scoring approach, taking only into account the nature of the image pairs rather than any features about kin information. To illustrate our motives, we provide classification results utilizing a simple scoring method based on the image similarity of both images of a kinship pair. Using simply the distance of the chrominance averages of the images in the Lab color space without any training or using any specific kin features, we achieve performance comparable to state-of-the-art methods. We provide the source code to prove the validity of our claims and ensure the repeatability of our experiments.

KW - kinship verification

KW - face recognition

KW - biometrics

U2 - 10.1109/TPAMI.2016.2522416

DO - 10.1109/TPAMI.2016.2522416

M3 - Article

VL - 38

SP - 2342

EP - 2344

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 11

ER -