IEEE Trans Image Process. 2018 Jan;27(1):293-303. doi: 10.1109/TIP.2017.2756450. Epub 2017 Sep 25.
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.
深度卷积神经网络最近在不受控制的环境下的困难人脸识别问题中被证明非常有效。为了训练这样的网络，需要有包含数百万张标记图像的非常大的训练集。对于某些应用，例如近红外（NIR）人脸识别，这样的大型训练数据集不可用，并且难以收集。在本文中，我们提出了一种通过在给定数据集中组合真实人脸图像来生成非常大的合成图像训练集的方法。我们表明，这种方法可以从少至 10000 张训练图像中学习到模型，其性能与从 500000 张图像中训练的模型相当。使用我们的方法，我们还在 CASIA NIR-VIS2.0 异构人脸识别数据集上获得了最先进的结果。
