ViTNT-FIQA: Training-Free Face Image Quality Assessment with Vision Transformers

TL;DR

ViTNT-FIQA introduces a training-free face image quality assessment method using the stability of patch embeddings in Vision Transformers, enabling efficient and effective quality scoring without additional training.

Contribution

It proposes a novel, training-free approach that measures patch embedding stability across ViT blocks for face image quality assessment, requiring only a single forward pass.

Findings

Achieves competitive performance on eight benchmarks.

Requires only one forward pass without backpropagation.

Maintains computational efficiency and broad applicability.

Abstract

Face Image Quality Assessment (FIQA) is essential for reliable face recognition systems. Current approaches primarily exploit only final-layer representations, while training-free methods require multiple forward passes or backpropagation. We propose ViTNT-FIQA, a training-free approach that measures the stability of patch embedding evolution across intermediate Vision Transformer (ViT) blocks. We demonstrate that high-quality face images exhibit stable feature refinement trajectories across blocks, while degraded images show erratic transformations. Our method computes Euclidean distances between L2-normalized patch embeddings from consecutive transformer blocks and aggregates them into image-level quality scores. We empirically validate this correlation on a quality-labeled synthetic dataset with controlled degradation levels. Unlike existing training-free approaches, ViTNT-FIQA requires only a single forward pass without backpropagation or architectural modifications. Through extensive evaluation on eight benchmarks (LFW, AgeDB-30, CFP-FP, CALFW, Adience, CPLFW, XQLFW, IJB-C), we show that ViTNT-FIQA achieves competitive performance with state-of-the-art methods while maintaining computational efficiency and immediate applicability to any pre-trained ViT-based face recognition model.