Head poses and grimaces: Challenges for automated face
identification algorithms?

J 2024

Head poses and grimaces: Challenges for automated face identification algorithms?

URBANOVÁ, Petra, Tomáš GOLDMANN, Dominik ČERNÝ and Martin DRAHANSKÝ

Basic information

Original name

Head poses and grimaces: Challenges for automated face identification algorithms?

Authors

URBANOVÁ, Petra (203 Czech Republic, guarantor, belonging to the institution), Tomáš GOLDMANN (203 Czech Republic), Dominik ČERNÝ (203 Czech Republic, belonging to the institution) and Martin DRAHANSKÝ (203 Czech Republic, belonging to the institution)

Edition

SCIENCE & JUSTICE, ENGLAND, ELSEVIER SCI LTD, 2024, 1355-0306

Other information

Language

English

Type of outcome

Článek v odborném periodiku

Field of Study

10201 Computer sciences, information science, bioinformatics

Country of publisher

United Kingdom of Great Britain and Northern Ireland

Confidentiality degree

není předmětem státního či obchodního tajemství

References:

URL

Impact factor

Impact factor: 1.900 in 2022

Organization unit

Faculty of Science

DOI

http://dx.doi.org/10.1016/j.scijus.2024.06.002

UT WoS

001262494100001

Keywords in English

Forensic image identification; Automated algorithms; Head pose; Facial expressions

Abstract

V originále

In today’s biometric and commercial settings, state-of-the-art image processing relies solely on artificial intelligence and machine learning which provides a high level of accuracy. However, these principles are deeply rooted in abstract, complex “black-box systems”. When applied to forensic image identification, concerns about transparency and accountability emerge. This study explores the impact of two challenging factors in automated facial identification: facial expressions and head poses. The sample comprised 3D faces with nine prototype expressions, collected from 41 participants (13 males, 28 females) of European descent aged 19.96 to 50.89 years. Pre-processing involved converting 3D models to 2D color images (256 × 256 px). Probes included a set of 9 images per individual with head poses varying by 5° in both left-to-right (yaw) and up-and-down (pitch) directions for neutral expressions. A second set of 3,610 images per individual covered viewpoints in 5° increments from −45° to 45° for head movements and different facial expressions, forming the targets. Pair-wise comparisons using ArcFace, a state-of-the-art face identification algorithm yielded 54,615,690 dissimilarity scores. Results indicate that minor head deviations in probes have minimal impact. However, the performance diminished as targets deviated from the frontal position. Right-to-left movements were less influential than up and down, with downward pitch showing less impact than upward movements. The lowest accuracy was for upward pitch at 45°. Dissimilarity scores were consistently higher for males than for females across all studied factors. The performance particularly diverged in upward movements, starting at 15°. Among tested facial expressions, happiness and contempt performed best, while disgust exhibited the lowest AUC values.

Links

MUNI/A/1597/2023, interní kód MU

Name: Rozvoj automatizace při zpracování obrazových dat ve forenzní a aplikované antropologii

Investor: Masaryk University, Automated Image Processing in Forensic and Applied Anthropology

VB02000062, research and development project

Name: Detekce a identifikace osob v davu na základě letecké prospekce

Investor: Ministry of the Interior of the CR, Person Detection and Identification in Crowds by Aerial Footage

Citovat

URBANOVÁ, Petra, Tomáš GOLDMANN, Dominik ČERNÝ and Martin DRAHANSKÝ. Head poses and grimaces: Challenges for automated face identification algorithms? SCIENCE & JUSTICE. ENGLAND: ELSEVIER SCI LTD, 2024, vol. 64, No 4, p. 421-442. ISSN 1355-0306. Available from: https://dx.doi.org/10.1016/j.scijus.2024.06.002.

@article{2413671,
   author = {Urbanová, Petra and Goldmann, Tomáš and Černý, Dominik and Drahanský, Martin},
   article_location = {ENGLAND},
   article_number = {4},
   doi = {http://dx.doi.org/10.1016/j.scijus.2024.06.002},
   keywords = {Forensic image identification; Automated algorithms; Head pose; Facial expressions},
   language = {eng},
   issn = {1355-0306},
   journal = {SCIENCE & JUSTICE},
   title = {Head poses and grimaces: Challenges for automated face identification algorithms?},
   url = {https://www.sciencedirect.com/science/article/pii/S1355030624000522?via%3Dihub},
   volume = {64},
   year = {2024}
}

TY  - JOUR
ID  - 2413671
AU  - Urbanová, Petra - Goldmann, Tomáš - Černý, Dominik - Drahanský, Martin
PY  - 2024
TI  - Head poses and grimaces: Challenges for automated face identification algorithms?
JF  - SCIENCE & JUSTICE
VL  - 64
IS  - 4
SP  - 421-442
EP  - 421-442
PB  - ELSEVIER SCI LTD
SN  - 13550306
KW  - Forensic image identification
KW  - Automated algorithms
KW  - Head pose
KW  - Facial expressions
UR  - https://www.sciencedirect.com/science/article/pii/S1355030624000522?via%3Dihub
N2  - In today’s biometric and commercial settings, state-of-the-art image processing relies solely on artificial intelligence and machine learning which provides a high level of accuracy. However, these principles are deeply rooted in abstract, complex “black-box systems”. When applied to forensic image identification, concerns about transparency and accountability emerge. This study explores the impact of two challenging factors in automated facial identification: facial expressions and head poses. The sample comprised 3D faces with nine prototype expressions, collected from 41 participants (13 males, 28 females) of European descent aged 19.96 to 50.89 years. Pre-processing involved converting 3D models to 2D color images (256 × 256 px). Probes included a set of 9 images per individual with head poses varying by 5° in both left-to-right (yaw) and up-and-down (pitch) directions for neutral expressions. A second set of 3,610 images per individual covered viewpoints in 5° increments from −45° to 45° for head movements and different facial expressions, forming the targets. Pair-wise comparisons using ArcFace, a state-of-the-art face identification algorithm yielded 54,615,690 dissimilarity scores. Results indicate that minor head deviations in probes have minimal impact. However, the performance diminished as targets deviated from the frontal position. Right-to-left movements were less influential than up and down, with downward pitch showing less impact than upward movements. The lowest accuracy was for upward pitch at 45°. Dissimilarity scores were consistently higher for males than for females across all studied factors. The performance particularly diverged in upward movements, starting at 15°. Among tested facial expressions, happiness and contempt performed best, while disgust exhibited the lowest AUC values.
ER  -

URBANOVÁ, Petra, Tomáš GOLDMANN, Dominik ČERNÝ and Martin DRAHANSKÝ. Head poses and grimaces: Challenges for automated face identification algorithms? \textit{SCIENCE \&{} JUSTICE}. ENGLAND: ELSEVIER SCI LTD, 2024, vol.~64, No~4, p.~421-442. ISSN~1355-0306. Available from: https://dx.doi.org/10.1016/j.scijus.2024.06.002.

Detailed Information on Publication Record