# Boosting Unconstrained Face Recognition with Targeted Style Adversary

arXiv: https://arxiv.org/abs/2408.07642 · HTML: https://mlsanity.com/paper/2408.07642/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 46 of its 75 references that are in the map.
- First posted on arXiv 14 Aug 2024.

## Builds on (most cited first, 30 of 46)

- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Towards Deep Learning Models Resistant to Adversarial Attacks](https://mlsanity.com/paper/1706.06083/) (2017)
- [Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization](https://mlsanity.com/paper/1703.06868/) (2017)
- [Instance Normalization: The Missing Ingredient for Fast Stylization](https://mlsanity.com/paper/1607.08022/) (2016)
- [CyCADA: Cycle-Consistent Adversarial Domain Adaptation](https://mlsanity.com/paper/1711.03213/) (2017)
- [CosFace: Large Margin Cosine Loss for Deep Face Recognition](https://mlsanity.com/paper/1801.09414/) (2018)
- [MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognition](https://mlsanity.com/paper/1607.08221/) (2016)
- [Domain Generalization with MixStyle](https://mlsanity.com/paper/2104.02008/) (2021)
- [Generalizing to Unseen Domains via Adversarial Data Augmentation](https://mlsanity.com/paper/1805.12018/) (2018)
- [AdaFace: Quality Adaptive Margin for Face Recognition](https://mlsanity.com/paper/2204.00964/) (2022)
- [WIDER FACE: A Face Detection Benchmark](https://mlsanity.com/paper/1511.06523/) (2015)
- [Learning to Learn Single Domain Generalization](https://mlsanity.com/paper/2003.13216/) (2020)
- [Data Augmentation Generative Adversarial Networks](https://mlsanity.com/paper/1711.04340/) (2017)
- [Generate To Adapt: Aligning Domains using Generative Adversarial Networks](https://mlsanity.com/paper/1704.01705/) (2017)
- [Efficient Decision-based Black-box Adversarial Attacks on Face Recognition](https://mlsanity.com/paper/1904.04433/) (2019)
- [Probabilistic Face Embeddings](https://mlsanity.com/paper/1904.09658/) (2019)
- [FSRNet: End-to-End Learning Face Super-Resolution with Facial Priors](https://mlsanity.com/paper/1711.10703/) (2017)
- [AdvHat: Real-world adversarial attack on ArcFace Face ID system](https://mlsanity.com/paper/1908.08705/) (2019)
- [Deep Learning for Single Image Super-Resolution: A Brief Review](https://mlsanity.com/paper/1808.03344/) (2018)
- [Exact Feature Distribution Matching for Arbitrary Style Transfer and Domain Generalization](https://mlsanity.com/paper/2203.07740/) (2022)
- [Adv-Makeup: A New Imperceptible and Transferable Attack on Face Recognition](https://mlsanity.com/paper/2105.03162/) (2021)
- [Sample and Computation Redistribution for Efficient Face Detection](https://mlsanity.com/paper/2105.04714/) (2021)
- [Adversarially Adaptive Normalization for Single Domain Generalization](https://mlsanity.com/paper/2106.01899/) (2021)
- [Beyond Face Rotation: Global and Local Perception GAN for Photorealistic and Identity Preserving Frontal View Synthesis](https://mlsanity.com/paper/1704.04086/) (2017)
- [Progressive Face Super-Resolution via Attention to Facial Landmark](https://mlsanity.com/paper/1908.08239/) (2019)
- [Adversarial Style Augmentation for Domain Generalized Urban-Scene Segmentation](https://mlsanity.com/paper/2207.04892/) (2022)
- [Low-Resolution Face Recognition](https://mlsanity.com/paper/1811.08965/) (2018)
- [AdvFaces: Adversarial Face Synthesis](https://mlsanity.com/paper/1908.05008/) (2019)
- [Deep Face Super-Resolution with Iterative Collaboration between Attentive Recovery and Landmark Estimation](https://mlsanity.com/paper/2003.13063/) (2020)

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

- [WaFusion: A Wavelet-Enhanced Diffusion Framework for Face Morph Generation](https://mlsanity.com/paper/2507.12493/) (2025)

## Abstract (arXiv)

While deep face recognition models have demonstrated remarkable performance, they often struggle on the inputs from domains beyond their training data. Recent attempts aim to expand the training set by relying on computationally expensive and inherently challenging image-space augmentation of image generation modules. In an orthogonal direction, we present a simple yet effective method to expand the training data by interpolating between instance-level feature statistics across labeled and unlabeled sets. Our method, dubbed Targeted Style Adversary (TSA), is motivated by two observations: (i) the input domain is reflected in feature statistics, and (ii) face recognition model performance is influenced by style information. Shifting towards an unlabeled style implicitly synthesizes challenging training instances. We devise a recognizability metric to constraint our framework to preserve the inherent identity-related information of labeled instances. The efficacy of our method is demonstrated through evaluations on unconstrained benchmarks, outperforming or being on par with its competitors while offering nearly a 70% improvement in training speed and 40% less memory consumption.