Real-Time 3D Face Identification from a Depth Camera
EURECOM-France
21st International Conference on Pattern Recognition (ICPR 2012), 2012
@inproceedings{EURECOM3764,
year={2012},
title={R}eal-time 3{D} face identification from a depth camera},
author={M}in, {R}ui and {C}hoi, {J}ongmoo and {M}edioni, {G}{‘e}rard and {D}ugelay, {J}ean-{L}uc},
booktitle={ICPR} 2012, 21st {I}nternational {C}onference on {P}attern {R}ecognition, {N}ovember 11-15, 2012, {T}sukuba {I}nternational {C}ongress {C}enter, {T}sukuba {S}cience {C}ity, {J}apan},
address={T}sukuba, {JAPAN},
month={11},
url={http://www.eurecom.fr/publication/3764}
}
We present a real-time 3D face identification system using a consumer level depth camera (PrimeSensor). Our system takes a noisy sequence as input and produces reliable identification. Instead of registering a probe to all instances in the database, we propose to only register it with several intermediate references, which considerably reduces processing, while preserving the recognition rate. The presented system routinely achieves 100% identification rate when matching a (0.5-4 seconds) video sequence, and 97.9% for single frame recognition. These numbers refer to a real-world dataset of 20 people. The methodology extends directly to very large datasets. The process runs at 20fps on an off the shelf laptop.
February 14, 2013 by hgpu
