Does Head Pose Correction Improve Biometric Facial Recognition?

TL;DR

This study evaluates whether AI-driven head pose correction and image restoration techniques can enhance biometric facial recognition accuracy in challenging real-world conditions.

Contribution

It systematically assesses multiple restoration methods and identifies effective combinations that improve recognition performance.

Findings

Naive application of restoration techniques degrades accuracy.

Selective use of CFR-GAN and CodeFormer improves recognition results.

Different restoration methods have varying impacts on recognition performance.

Abstract

Biometric facial recognition models often demonstrate significant decreases in accuracy when processing real-world images, often characterized by poor quality, non-frontal subject poses, and subject occlusions. We investigate whether targeted, AI-driven, head-pose correction and image restoration can improve recognition accuracy. Using a model-agnostic, large-scale, forensic-evaluation pipeline, we assess the impact of three restoration approaches: 3D reconstruction (NextFace), 2D frontalization (CFR-GAN), and feature enhancement (CodeFormer). We find that naive application of these techniques substantially degrades facial recognition accuracy. However, we also find that selective application of CFR-GAN combined with CodeFormer yields meaningful improvements.