Reliable gait recognition using 3D reconstructions and random forests: An anthropometric approach

Martin Sandau, Rikke V. Heimbürger, Karl Erik Jensen, Thomas B. Moeslund, Henrik Aanæs, Tine Alkjær, Erik B. Simonsen

Research output: Contribution to journalJournal articleResearchpeer-review

2 Citations (Scopus)

Abstract

Photogrammetric measurements of bodily dimensions and analysis of gait patterns in CCTV are important tools in forensic investigations but accurate extraction of the measurements are challenging. This study tested whether manual annotation of the joint centers on 3D reconstructions could provide reliable recognition. Sixteen participants performed normal walking where 3D reconstructions were obtained continually. Segment lengths and kinematics from the extremities were manually extracted by eight expert observers. The results showed that all the participants were recognized, assuming the same expert annotated the data. Recognition based on data annotated by different experts was less reliable achieving 72.6% correct recognitions as some parameters were heavily affected by interobserver variability. This study verified that 3D reconstructions are feasible for forensic gait analysis as an improved alternative to conventional CCTV. However, further studies are needed to account for the use of different clothing, field conditions, etc.
Original languageEnglish
JournalJournal of Forensic Sciences
Volume61
Issue number3
Pages (from-to)637-648
Number of pages12
ISSN0022-1198
DOIs
Publication statusPublished - 6 Jan 2016

Fingerprint

Gait
Clothing
Observer Variation
Biomechanical Phenomena
Walking
Extremities
Forests

Cite this

Sandau, Martin ; Heimbürger, Rikke V. ; Jensen, Karl Erik ; Moeslund, Thomas B. ; Aanæs, Henrik ; Alkjær, Tine ; Simonsen, Erik B. / Reliable gait recognition using 3D reconstructions and random forests : An anthropometric approach. In: Journal of Forensic Sciences. 2016 ; Vol. 61, No. 3. pp. 637-648.
@article{c615532f2cee4961b8a7cf421e70370b,
title = "Reliable gait recognition using 3D reconstructions and random forests: An anthropometric approach",
abstract = "Photogrammetric measurements of bodily dimensions and analysis of gait patterns in CCTV are important tools in forensic investigations but accurate extraction of the measurements are challenging. This study tested whether manual annotation of the joint centers on 3D reconstructions could provide reliable recognition. Sixteen participants performed normal walking where 3D reconstructions were obtained continually. Segment lengths and kinematics from the extremities were manually extracted by eight expert observers. The results showed that all the participants were recognized, assuming the same expert annotated the data. Recognition based on data annotated by different experts was less reliable achieving 72.6{\%} correct recognitions as some parameters were heavily affected by interobserver variability. This study verified that 3D reconstructions are feasible for forensic gait analysis as an improved alternative to conventional CCTV. However, further studies are needed to account for the use of different clothing, field conditions, etc.",
author = "Martin Sandau and Heimb{\"u}rger, {Rikke V.} and Jensen, {Karl Erik} and Moeslund, {Thomas B.} and Henrik Aan{\ae}s and Tine Alkj{\ae}r and Simonsen, {Erik B.}",
year = "2016",
month = "1",
day = "6",
doi = "10.1111/1556-4029.13015",
language = "English",
volume = "61",
pages = "637--648",
journal = "Journal of Forensic Sciences",
issn = "0022-1198",
publisher = "Wiley-Blackwell",
number = "3",

}

Reliable gait recognition using 3D reconstructions and random forests : An anthropometric approach. / Sandau, Martin; Heimbürger, Rikke V.; Jensen, Karl Erik; Moeslund, Thomas B.; Aanæs, Henrik; Alkjær, Tine; Simonsen, Erik B.

In: Journal of Forensic Sciences, Vol. 61, No. 3, 06.01.2016, p. 637-648.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Reliable gait recognition using 3D reconstructions and random forests

T2 - An anthropometric approach

AU - Sandau, Martin

AU - Heimbürger, Rikke V.

AU - Jensen, Karl Erik

AU - Moeslund, Thomas B.

AU - Aanæs, Henrik

AU - Alkjær, Tine

AU - Simonsen, Erik B.

PY - 2016/1/6

Y1 - 2016/1/6

N2 - Photogrammetric measurements of bodily dimensions and analysis of gait patterns in CCTV are important tools in forensic investigations but accurate extraction of the measurements are challenging. This study tested whether manual annotation of the joint centers on 3D reconstructions could provide reliable recognition. Sixteen participants performed normal walking where 3D reconstructions were obtained continually. Segment lengths and kinematics from the extremities were manually extracted by eight expert observers. The results showed that all the participants were recognized, assuming the same expert annotated the data. Recognition based on data annotated by different experts was less reliable achieving 72.6% correct recognitions as some parameters were heavily affected by interobserver variability. This study verified that 3D reconstructions are feasible for forensic gait analysis as an improved alternative to conventional CCTV. However, further studies are needed to account for the use of different clothing, field conditions, etc.

AB - Photogrammetric measurements of bodily dimensions and analysis of gait patterns in CCTV are important tools in forensic investigations but accurate extraction of the measurements are challenging. This study tested whether manual annotation of the joint centers on 3D reconstructions could provide reliable recognition. Sixteen participants performed normal walking where 3D reconstructions were obtained continually. Segment lengths and kinematics from the extremities were manually extracted by eight expert observers. The results showed that all the participants were recognized, assuming the same expert annotated the data. Recognition based on data annotated by different experts was less reliable achieving 72.6% correct recognitions as some parameters were heavily affected by interobserver variability. This study verified that 3D reconstructions are feasible for forensic gait analysis as an improved alternative to conventional CCTV. However, further studies are needed to account for the use of different clothing, field conditions, etc.

U2 - 10.1111/1556-4029.13015

DO - 10.1111/1556-4029.13015

M3 - Journal article

VL - 61

SP - 637

EP - 648

JO - Journal of Forensic Sciences

JF - Journal of Forensic Sciences

SN - 0022-1198

IS - 3

ER -