# Frankenstein: Learning Deep Face Representations using Small Data

arXiv: https://arxiv.org/abs/1603.06470 · HTML: https://mlsanity.com/paper/1603.06470/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Mar 2016.

## Built on by (most cited first, 7 of 7)

- [TANGOS: Regularizing Tabular Neural Networks through Gradient Orthogonalization and Specialization](https://mlsanity.com/paper/2303.05506/) (2023)
- [An Efficient Training Approach for Very Large Scale Face Recognition](https://mlsanity.com/paper/2105.10375/) (2021)
- [A Comparative Study of Deep Learning Classification Methods on a Small Environmental Microorganism Image Dataset (EMDS-6): from Convolutional Neural Networks to Visual Transformers](https://mlsanity.com/paper/2107.07699/) (2021)
- [Residual Tensor Train: A Quantum-inspired Approach for Learning Multiple Multilinear Correlations](https://mlsanity.com/paper/2108.08659/) (2021)
- [On the Effectiveness of Neural Ensembles for Image Classification with Small Datasets](https://mlsanity.com/paper/2111.14493/) (2021)
- [ScaleNet: An Unsupervised Representation Learning Method for Limited Information](https://mlsanity.com/paper/2310.02386/) (2023)
- [AttFC: Attention Fully-Connected Layer for Large-Scale Face Recognition with One GPU](https://mlsanity.com/paper/2503.06839/) (2025)

## Abstract (arXiv)

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 datasets are not publicly available and difficult to collect. In this work, we propose a method to generate very large training datasets of synthetic images by compositing real face images in a given dataset. 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 dataset.