AUTOMATSKO GENERISANJE SKUPA PODATAKA ZA TRENIRANJE MODELA ZA AUTOMATSKO PREPOZNAVANJE OSOBE NA SLICI

Noemi Sabadoš

doi:10.24867/12BE40Sabados

Noemi Sabadoš

DOI: https://doi.org/10.24867/12BE40Sabados

Ključne reči: skup podataka, generisanje skupa podataka, skup podataka o licu, skup podataka o slavnim ličnostima, detekcija atributa lica

Apstrakt

Kvalitetni skupovi podataka su retko dostupni, što otežava treniranje kvalitetnih modela mašinskog učenja. Cilj ovog rada je generisanje skupa podataka koji bi mogao biti korišćen za obuku modela mašinskog učenja čiji bi cilj bio da uporedi slike ljudi sa slikama poznatih ličnosti i prepozna sličnosti u izgledu. Generisanje skupa podataka se vrši automatski, na osnovu unapred zadatih kriterijuma. Svrha kriterijuma je da učine skup podataka raznovrsnim, sa ciljem poboljšanja generalizacije nad njime obučenih modela. Podržani kriterijumi uključivanja slika u skup podataka su sledeći: (1) da li se osoba smeši ili ne, (2) specifikacija smera pogleda, (3) činjenica da li osoba ima: bradu, šiške, kosu, (4) da li osoba nosi naočare ili kapu i (5) da li osoba rukom zaklanja lice. Prikupljeni skup podataka sadrži slike o 100 glumca, koje su skinute sa interneta prema zadatim kriterijumima.

Reference

[1] Huang, Gary B., et al. "Labeled faces in the wild: A database for studying face recognition in unconstrained environments." 2008.
[2] Berg, Tamara L., et al. "Names and faces in the news." Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004.. Vol. 2. IEEE, 2004.
[3] Liu, Ziwei, et al. "Deep learning face attributes in the wild." Proceedings of the IEEE international conference on computer vision. 2015.
[4] Sun, Yi, Xiaogang Wang, and Xiaoou Tang. "Deep learning face representation from predicting 10,000 classes." Proceedings of the IEEE conference on computer vision and pattern recognition. 2014.
[5] Bansal, Ankan, et al. "Umdfaces: An annotated face dataset for training deep networks." 2017 IEEE International Joint Conference on Biometrics (IJCB). IEEE, 2017.
[6] Déniz, Oscar, et al. "Face recognition using histograms of oriented gradients." Pattern recognition letters 32.12 (2011): 1598-1603.
[7] Amos, Brandon, Bartosz Ludwiczuk, and Mahadev Satyanarayanan. "Openface: A general-purpose face recognition library with mobile applications." CMU School of Computer Science 6.2 (2016).
[8] Soukupová, Tereza, and Jan Cech. "Real-time eye blink detection using facial landmarks." 21st Computer Vision Winter Workshop. 2016.
[9] Chen, Junkai, et al. "Smile detection in the wild with deep convolutional neural networks." Machine vision and applications 28.1-2 (2017): 173-183.
[10] Zhang, Kaihao, et al. "Facial smile detection based on deep learning features." 2015 3rd IAPR Asian Conference on Pattern Recognition (ACPR). IEEE, 2015.
[11] Gourier, Nicolas, Daniela Hall, and James L. Crowley. "Estimating face orientation from robust detection of salient facial features." ICPR International Workshop on Visual Observation of Deictic Gestures. 2004.
[12] He, Kaiming, et al. "Mask r-cnn." Proc. of the IEEE international conference on computer vision. 2017.
[13] Kuznetsova, Alina, et al. "The open images dataset v4: Unified image classification, object detection, and visual relationship detection at scale." arXiv preprint arXiv:1811.00982 (2018).
[14] Simon, Tomas, et al. "Hand keypoint detection in single images using multiview bootstrapping." Proc. of the IEEE conference on Computer Vision and Pattern Recognition. 2017.