Manually annotated characteristic descriptors: Measurability and variability

Chris Zeinstra; Raymond Veldhuis; Luuk Spreeuwers; Arnout Ruifrok

doi:10.1109/IWBF.2017.7935095

Manually annotated characteristic descriptors: Measurability and variability

Chris Zeinstra, Raymond Veldhuis, Luuk Spreeuwers, Arnout Ruifrok

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

1 Citation (Scopus)

11 Downloads (Pure)

Abstract

In this paper we study the measurability and variability of manually annotated characteristic descriptors on a forensic relevant face dataset. Characteristic descriptors are facial features (landmarks, shapes, etc.) that can be used during forensic case work. With respect to measurability, we observe that a significant proportion cannot be determined in images representative of forensic case work. Landmarks, closed and open shapes, and other forensic facial features show mostly that the variability depends on the image quality. Up to 50% of all considered evidential values are either positively or negatively influenced by annotator variability. However, when considering images with the lowest quality, we found that more than 70% of the evidential value intervals in principle could yield the wrong conclusion.

Original language	English
Title of host publication	2017 5th International Workshop on Biometrics and Forensics (IWBF 2017)
Publisher	IEEE
ISBN (Electronic)	9781509057917
ISBN (Print)	9781509057924
DOIs	https://doi.org/10.1109/IWBF.2017.7935095
Publication status	Published - 26 May 2017
Event	5th International Workshop on Biometrics and Forensics, IWBF 2017 - Coventry, United Kingdom Duration: 4 Apr 2017 → 5 Apr 2017 Conference number: 5 https://warwick.ac.uk/fac/sci/dcs/people/chang-tsun_li/iwbf2017/

Conference

Conference	5th International Workshop on Biometrics and Forensics, IWBF 2017
Abbreviated title	IWBF 2017
Country/Territory	United Kingdom
City	Coventry
Period	4/04/17 → 5/04/17
Internet address	https://warwick.ac.uk/fac/sci/dcs/people/chang-tsun_li/iwbf2017/

Keywords

FISWG
Forensic facial features
Manual annotation
Measurability
Variability

Access to Document

10.1109/IWBF.2017.7935095

Cite this

@inproceedings{b22a78e0413249798df943f8fd229e56,

title = "Manually annotated characteristic descriptors: Measurability and variability",

abstract = "In this paper we study the measurability and variability of manually annotated characteristic descriptors on a forensic relevant face dataset. Characteristic descriptors are facial features (landmarks, shapes, etc.) that can be used during forensic case work. With respect to measurability, we observe that a significant proportion cannot be determined in images representative of forensic case work. Landmarks, closed and open shapes, and other forensic facial features show mostly that the variability depends on the image quality. Up to 50% of all considered evidential values are either positively or negatively influenced by annotator variability. However, when considering images with the lowest quality, we found that more than 70% of the evidential value intervals in principle could yield the wrong conclusion.",

keywords = "FISWG, Forensic facial features, Manual annotation, Measurability, Variability",

author = "Chris Zeinstra and Raymond Veldhuis and Luuk Spreeuwers and Arnout Ruifrok",

year = "2017",

month = may,

day = "26",

doi = "10.1109/IWBF.2017.7935095",

language = "English",

isbn = "9781509057924",

booktitle = "2017 5th International Workshop on Biometrics and Forensics (IWBF 2017)",

publisher = "IEEE",

address = "United States",

note = "5th International Workshop on Biometrics and Forensics, IWBF 2017, IWBF 2017 ; Conference date: 04-04-2017 Through 05-04-2017",

url = "https://warwick.ac.uk/fac/sci/dcs/people/chang-tsun_li/iwbf2017/",

}

Zeinstra, C , Veldhuis, R , Spreeuwers, L & Ruifrok, A 2017, Manually annotated characteristic descriptors: Measurability and variability. in 2017 5th International Workshop on Biometrics and Forensics (IWBF 2017) ., 7935095, IEEE, 5th International Workshop on Biometrics and Forensics, IWBF 2017, Coventry, United Kingdom, 4/04/17. https://doi.org/10.1109/IWBF.2017.7935095

TY - GEN

T1 - Manually annotated characteristic descriptors

T2 - 5th International Workshop on Biometrics and Forensics, IWBF 2017

AU - Zeinstra, Chris

AU - Veldhuis, Raymond

AU - Spreeuwers, Luuk

AU - Ruifrok, Arnout

N1 - Conference code: 5

PY - 2017/5/26

Y1 - 2017/5/26

N2 - In this paper we study the measurability and variability of manually annotated characteristic descriptors on a forensic relevant face dataset. Characteristic descriptors are facial features (landmarks, shapes, etc.) that can be used during forensic case work. With respect to measurability, we observe that a significant proportion cannot be determined in images representative of forensic case work. Landmarks, closed and open shapes, and other forensic facial features show mostly that the variability depends on the image quality. Up to 50% of all considered evidential values are either positively or negatively influenced by annotator variability. However, when considering images with the lowest quality, we found that more than 70% of the evidential value intervals in principle could yield the wrong conclusion.

AB - In this paper we study the measurability and variability of manually annotated characteristic descriptors on a forensic relevant face dataset. Characteristic descriptors are facial features (landmarks, shapes, etc.) that can be used during forensic case work. With respect to measurability, we observe that a significant proportion cannot be determined in images representative of forensic case work. Landmarks, closed and open shapes, and other forensic facial features show mostly that the variability depends on the image quality. Up to 50% of all considered evidential values are either positively or negatively influenced by annotator variability. However, when considering images with the lowest quality, we found that more than 70% of the evidential value intervals in principle could yield the wrong conclusion.

KW - FISWG

KW - Forensic facial features

KW - Manual annotation

KW - Measurability

KW - Variability

UR - http://www.scopus.com/inward/record.url?scp=85021801338&partnerID=8YFLogxK

U2 - 10.1109/IWBF.2017.7935095

DO - 10.1109/IWBF.2017.7935095

M3 - Conference contribution

AN - SCOPUS:85021801338

SN - 9781509057924

BT - 2017 5th International Workshop on Biometrics and Forensics (IWBF 2017)

PB - IEEE

Y2 - 4 April 2017 through 5 April 2017

ER -

Manually annotated characteristic descriptors: Measurability and variability

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this