Frankenstein: Learning Deep Face Representations Using Small Data

doi:10.1109/TIP.2017.2756450

Frankenstein: Learning Deep Face Representations Using Small Data

2018-01-01

82

OA

AI

Timothy M. Hospedales

J

Jakob Verbeek

DOI：10.1109/TIP.2017.2756450

Original

Save

View PDF

Abstract

En 中文

Deep convolutional neural networks have recently proven extremely effective for difficult face recognition problems in uncontrolled settings. To train such networks, very large training sets are needed with millions of labeled images. For some applications, such as near-infrared (NIR) face recognition, such large training data sets are not publicly available and difficult to collect. In this paper, we propose a method to generate very large training data sets of synthetic images by compositing real face images in a given data set. We show that this method enables to learn models from as few as 10 000 training images, which perform on par with models trained from 500 000 images. Using our approach, we also obtain state-of-the-art results on the CASIA NIR-VIS2.0 heterogeneous face recognition data set.

Keywords:

Face recognition

deep learning

small training data

Frankenstein: Learning Deep Face Representations Using Small Data

Abstract

AI Summary

Journal

Organization