DArFace: Deformation Aware Robustness for Low Quality Face Recognition

TL;DR

DArFace is a novel face recognition framework that improves robustness to low-quality images by modeling both global and local deformations during training, leading to better performance on challenging benchmarks.

Contribution

The paper introduces DArFace, which incorporates adversarially simulated local elastic deformations and a contrastive loss to enhance face recognition robustness without needing paired training samples.

Findings

Outperforms state-of-the-art on TinyFace, IJB-B, IJB-C datasets.

Significant gains from modeling local elastic deformations.

Effective in real-world low-quality face recognition scenarios.

Abstract

Facial recognition systems have achieved remarkable success by leveraging deep neural networks, advanced loss functions, and large-scale datasets. However, their performance often deteriorates in real-world scenarios involving low-quality facial images. Such degradations, common in surveillance footage or standoff imaging include low resolution, motion blur, and various distortions, resulting in a substantial domain gap from the high-quality data typically used during training. While existing approaches attempt to address robustness by modifying network architectures or modeling global spatial transformations, they frequently overlook local, non-rigid deformations that are inherently present in real-world settings. In this work, we introduce \textbf{DArFace}, a \textbf{D}eformation-\textbf{A}ware \textbf{r}obust \textbf{Face} recognition framework that enhances robustness to such degradations without requiring paired high- and low-quality training samples. Our method adversarially integrates both global transformations (e.g., rotation, translation) and local elastic deformations during training to simulate realistic low-quality conditions. Moreover, we introduce a contrastive objective to enforce identity consistency across different deformed views. Extensive evaluations on low-quality benchmarks including TinyFace, IJB-B, and IJB-C demonstrate that DArFace surpasses state-of-the-art methods, with significant gains attributed to the inclusion of local deformation modeling.