Reliable Gait Recognition Using 3D Reconstructions and Random Forests - An Anthropometric Approach

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt

Standard

Reliable Gait Recognition Using 3D Reconstructions and Random Forests - An Anthropometric Approach. / Christiansen, Martin Sandau; Heimbürger, Rikke V; Jensen, Karl E; Moeslund, Thomas B; Aanaes, Henrik; Alkjaer, Tine; Simonsen, Erik B.

I: Journal of Forensic Sciences, Bind 61, Nr. 3, 05.2016, s. 637-48.

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt

Harvard

Christiansen, MS, Heimbürger, RV, Jensen, KE, Moeslund, TB, Aanaes, H, Alkjaer, T & Simonsen, EB 2016, 'Reliable Gait Recognition Using 3D Reconstructions and Random Forests - An Anthropometric Approach', Journal of Forensic Sciences, bind 61, nr. 3, s. 637-48. https://doi.org/10.1111/1556-4029.13015

APA

Christiansen, M. S., Heimbürger, R. V., Jensen, K. E., Moeslund, T. B., Aanaes, H., Alkjaer, T., & Simonsen, E. B. (2016). Reliable Gait Recognition Using 3D Reconstructions and Random Forests - An Anthropometric Approach. Journal of Forensic Sciences, 61(3), 637-48. https://doi.org/10.1111/1556-4029.13015

Vancouver

Christiansen MS, Heimbürger RV, Jensen KE, Moeslund TB, Aanaes H, Alkjaer T o.a. Reliable Gait Recognition Using 3D Reconstructions and Random Forests - An Anthropometric Approach. Journal of Forensic Sciences. 2016 maj;61(3):637-48. https://doi.org/10.1111/1556-4029.13015

Author

Christiansen, Martin Sandau ; Heimbürger, Rikke V ; Jensen, Karl E ; Moeslund, Thomas B ; Aanaes, Henrik ; Alkjaer, Tine ; Simonsen, Erik B. / Reliable Gait Recognition Using 3D Reconstructions and Random Forests - An Anthropometric Approach. I: Journal of Forensic Sciences. 2016 ; Bind 61, Nr. 3. s. 637-48.

Bibtex

@article{e8869f15fef646abb479bd9b97681b75,

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 = "Christiansen, {Martin Sandau} and Heimb{\"u}rger, {Rikke V} and Jensen, {Karl E} and Moeslund, {Thomas B} and Henrik Aanaes and Tine Alkjaer and Simonsen, {Erik B}",

note = "{\textcopyright} 2016 American Academy of Forensic Sciences.",

year = "2016",

month = may,

doi = "10.1111/1556-4029.13015",

language = "English",

volume = "61",

pages = "637--48",

journal = "Journal of Forensic Sciences",

issn = "0022-1198",

publisher = "Wiley-Blackwell",

number = "3",

}

RIS

TY - JOUR

T1 - Reliable Gait Recognition Using 3D Reconstructions and Random Forests - An Anthropometric Approach

AU - Christiansen, Martin Sandau

AU - Heimbürger, Rikke V

AU - Jensen, Karl E

AU - Moeslund, Thomas B

AU - Aanaes, Henrik

AU - Alkjaer, Tine

AU - Simonsen, Erik B

PY - 2016/5

Y1 - 2016/5

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

C2 - 27122399

VL - 61

SP - 637

EP - 648

JO - Journal of Forensic Sciences

JF - Journal of Forensic Sciences

SN - 0022-1198

IS - 3

ER -

ID: 161560707