Improving Low-Resolution Face Recognition under Limited Data: How Synthetic Data Generation Can Close the Domain Gap

2026-08-06

AI Summary

Low-resolution faces are prevalent in surveillance scenarios, yet the scarcity of real-world low-resolution images and their corresponding high-resolution pairs significantly limits face recognition performance. This study systematically evaluates the effectiveness of various synthetic low-resolution data generation strategies—including interpolation-based degradation, Real-ESRGAN-style degradation, and learnable super-resolution frontends—on lightweight edge devices, while also examining the domain gap between synthetic and real low-resolution data. Experimental results demonstrate that more complex synthesis methods do not necessarily yield better recognition accuracy; notably, simple interpolation-based augmentation consistently outperforms the baseline of directly feeding low-resolution inputs. These findings underscore the critical importance of validating face recognition approaches on real low-resolution data rather than relying solely on synthetic proxies.

Abstract

Face Recognition (FR) systems in surveillance settings often encounter Low Resolution (LR) faces, those whose face region falls below the standard 112 $\times$ 112 input size. While labelled High Resolution (HR) training data is abundant, labelled native-LR data, and above all paired native LR/HR data, is scarce. One workaround is to synthesize LR data from the available HR faces, but how much synthesis effort is repaid in recognition accuracy remains unclear. We present a study of simple synthetic generation strategies for a compact, edge device-oriented face recognition system, spanning interpolation-based degradation, knowledge distillation, a Prepended Domain Transformer (PDT), Real ESRGAN-style degradation, and a learned Super Resolution (SR) front-end with an identity-aware loss. We evaluate these strategies on synthetic cross-resolution face benchmarks (LFW, CFP-FP, AgeDB-30) and on TinyFace, a real-world native LR dataset, and expose a synthetic-real gap: the degradation setting that is optimal on synthetic benchmarks is not the one that is optimal on real LR. We find that more synthesis effort does not help monotonically: the learned SR front-end does not surpass a direct feed of the aligned LR image into a strong backbone, while simple interpolation augmentation of a compact backbone is the only synthesis that improves over its own baseline. We conclude that generative methods for LR face recognition must be validated on real LR and against a direct-feed baseline, and release our pipeline at https://idiap.ch/paper/synth-lrfr

Problem

• Low-Resolution Face Recognition
• Synthetic Data Generation
• Domain Gap
• Limited Data
• Face Recognition

Innovation

• Low-Resolution Face Recognition
• Synthetic Data Generation
• Domain Gap
• Interpolation Augmentation
• Edge Device