# Susceptibility to Image Resolution in Face Recognition and Trainings Strategies

arXiv: https://arxiv.org/abs/2107.03769 · HTML: https://mlsanity.com/paper/2107.03769/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 40 references that are in the map.
- First posted on arXiv 8 Jul 2021.

## Builds on (most cited first, 22 of 22)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Going Deeper with Convolutions](https://mlsanity.com/paper/1409.4842/) (2014)
- [FaceNet: A Unified Embedding for Face Recognition and Clustering](https://mlsanity.com/paper/1503.03832/) (2015)
- [CosFace: Large Margin Cosine Loss for Deep Face Recognition](https://mlsanity.com/paper/1801.09414/) (2018)
- [SphereFace: Deep Hypersphere Embedding for Face Recognition](https://mlsanity.com/paper/1704.08063/) (2017)
- [MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognition](https://mlsanity.com/paper/1607.08221/) (2016)
- [Cross-Age LFW: A Database for Studying Cross-Age Face Recognition in Unconstrained Environments](https://mlsanity.com/paper/1708.08197/) (2017)
- [Deep Plug-and-Play Super-Resolution for Arbitrary Blur Kernels](https://mlsanity.com/paper/1903.12529/) (2019)
- [Low-Resolution Face Recognition](https://mlsanity.com/paper/1811.08965/) (2018)
- [Exemplar Guided Face Image Super-Resolution without Facial Landmarks](https://mlsanity.com/paper/1906.07078/) (2019)
- [Low-resolution Face Recognition in the Wild via Selective Knowledge Distillation](https://mlsanity.com/paper/1811.09998/) (2018)
- [Super-Identity Convolutional Neural Network for Face Hallucination](https://mlsanity.com/paper/1811.02328/) (2018)
- [GroupFace: Learning Latent Groups and Constructing Group-based Representations for Face Recognition](https://mlsanity.com/paper/2005.10497/) (2020)
- [On Low-Resolution Face Recognition in the Wild: Comparisons and New Techniques](https://mlsanity.com/paper/1805.11529/) (2018)
- [Cross-Resolution Learning for Face Recognition](https://mlsanity.com/paper/1912.02851/) (2019)
- [Low Resolution Face Recognition Using a Two-Branch Deep Convolutional Neural Network Architecture](https://mlsanity.com/paper/1706.06247/) (2017)
- [SiGAN: Siamese Generative Adversarial Network for Identity-Preserving Face Hallucination](https://mlsanity.com/paper/1807.08370/) (2018)
- [Attribute-Guided Coupled GAN for Cross-Resolution Face Recognition](https://mlsanity.com/paper/1908.01790/) (2019)
- [Exploring Factors for Improving Low Resolution Face Recognition](https://mlsanity.com/paper/1907.10104/) (2019)
- [Identity-Aware Deep Face Hallucination via Adversarial Face Verification](https://mlsanity.com/paper/1909.08130/) (2019)
- [MagnifyMe: Aiding Cross Resolution Face Recognition via Identity Aware Synthesis](https://mlsanity.com/paper/1802.08057/) (2018)

## Built on by (most cited first, 1 of 1)

- [Tackling Face Verification Edge Cases: In-Depth Analysis and Human-Machine Fusion Approach](https://mlsanity.com/paper/2304.08134/) (2023)

## Abstract (arXiv)

Face recognition approaches often rely on equal image resolution for verifying faces on two images. However, in practical applications, those image resolutions are usually not in the same range due to different image capture mechanisms or sources. In this work, we first analyze the impact of image resolutions on face verification performance with a state-of-the-art face recognition model. For images synthetically reduced to 5,×,5 px resolution, the verification performance drops from 99.23% increasingly down to almost 55%. Especially for cross-resolution image pairs (one high- and one low-resolution image), the verification accuracy decreases even further. We investigate this behavior more in-depth by looking at the feature distances for every 2-image test pair. To tackle this problem, we propose the following two methods: 1) Train a state-of-the-art face-recognition model straightforwardly with 50% low-resolution images directly within each batch. 2) Train a siamese-network structure and add a cosine distance feature loss between high- and low-resolution features. Both methods show an improvement for cross-resolution scenarios and can increase the accuracy at very low resolution to approximately 70%. However, a disadvantage is that a specific model needs to be trained for every resolution pair. Thus, we extend the aforementioned methods by training them with multiple image resolutions at once. The performances for particular testing image resolutions are slightly worse, but the advantage is that this model can be applied to arbitrary resolution images and achieves an overall better performance (97.72% compared to 96.86%). Due to the lack of a benchmark for arbitrary resolution images for the cross-resolution and equal-resolution task, we propose an evaluation protocol for five well-known datasets, focusing on high, mid, and low-resolution images.