Learning Steadily: Accumulating Relative Point Margin Scores for Face Image Quality Assessment

Guray Ozgur; Tahar Chettaoui; Eduarda Caldeira; Marco Huber; Jan Niklas Kolf; Naser Damer; Fadi Boutros

cs.CVSep 15, 2026

Learning Steadily: Accumulating Relative Point Margin Scores for Face Image Quality Assessment

Authors: Guray Ozgur, Tahar Chettaoui, Eduarda Caldeira, Marco Huber, Jan Niklas Kolf, Naser Damer, Fadi Boutros

Organizations: Fraunhofer Institute for Computer Graphics Research IGD, Fraunhoferstr. 5, 64283 Darmstadt, Germany · Department of Computer Science at the TU Darmstadt, Karolinenpl. 5, 64289 Darmstadt, Germany

Abstract

Face Image Quality Assessment determines the suitability of captured face images for automated face recognition (FR), a critical capability for reliable biometric systems. Existing state-of-the-art FR-integrated FIQA methods suffer from temporal instability: as the feature space evolves during training, single-epoch quality estimates fluctuate, creating a moving target that undermines reliable quality prediction. We introduce CARPM-FIQA, a stabilization strategy for FR-integrated FIQA that accumulates relative point margin measurements, the ratio between intra-class compactness and inter-class separation, across the entire training trajectory rather than relying on single-epoch estimates. This cumulative averaging approach provides theoretically grounded advantages: reduced variance in quality estimates, improved mean squared error, and enhanced ranking stability with convergence guarantees as training progresses. Through controlled experiments on the SynFIQA dataset with labeled quality groups, we demonstrate that cumulative averaging achieves superior discriminative ability, and ablation studies across different training configurations confirm consistent improvements. Evaluated against twelve FIQA methods on eight challenging benchmarks with four FR models at two FMR thresholds, CARPM-FIQA places 4th (CARPM-FIQA(L)) and 6th (CARPM-FIQA(S)) of 17 compared methods by pAUC-EDC and AUC-EDC averaged across FR models and, after per-benchmark normalization, across benchmarks, staying within a few percent of the best method's normalized average for every FR model, providing a principled solution to training instability while maintaining the performance benefits of FR integration. More broadly, our work demonstrates that temporal aggregation strategies can stabilize training objectives in deep learning systems where target values inherently fluctuate due to evolving feature representations.

Figures & tables

Fig. 1: Normalized quality score distributions for 11 quality groups, each having 0.5M images, from 5.5M images of SynFIQA [ 1 ] . (Q0-Q9: degraded images from lowest to highest quality; Ref: reference images) comparing non-cumulative (top) and cumulative (bottom) methods. Scores are z-score normalized within each method. Color gradient from red to green indicates quality progression. Cohen’s d effect sizes between adjacent groups are reported. The cumulative method achieves higher average Cohen’s d (0.3112 vs. 0.2514) on the small protocol (ArcFace/ResNet50/CASIA-WebFace), demonstrating a better group separation.

Panel of Figure 2 — Fig. 2: High quality samples from identity #6987 across training epochs. Note how the top-ranked images in the non-cumulative approach (top) tend to stabilize in the cumulative approach (bottom), showing the benefit of averaging quality scores over time.

Panel of Figure 3 — Fig. 3: Comparison of cumulative and non-cumulative quality estimation. (a) EDC curves and (b) AUC-EDC trends show cumulative averaging improves estimates consistently, while non-cumulative degrades after epoch 20. (c, d) Quality target distributions highlight non-cumulative compression versus cumulative stability.

Metric	FR	Quality Metric	Adience		AgeDB-30		CFP-FP		LFW		CALFW		CPLFW		XQLFW		Avg. norm.
			[ 18 ]		[ 15 ]		[ 16 ]		[ 14 ]		[ 17 ]		[ 19 ]		[ 20 ]
			$1e{-3}$	$1e{-4}$	$1e{-3}$	$1e{-4}$	$1e{-3}$	$1e{-4}$	$1e{-3}$	$1e{-4}$	$1e{-3}$	$1e{-4}$	$1e{-3}$	$1e{-4}$	$1e{-3}$	$1e{-4}$
AUC-EDC	ArcFace [ 22 ]	non-cumulative	23.938	56.259	17.593	22.133	6.792	10.693	1.688	2.155	57.384	61.981	39.313	56.546	260.345	312.863	1.052
	ArcFace [ 22 ]	cumulative	22.070	44.323	17.897	22.728	6.791	10.743	1.953	2.573	60.759	63.547	41.281	57.975	218.855	262.735	1.040
	ElasticFace [ 23 ]	non-cumulative	26.439	49.392	17.313	18.149	7.707	11.052	1.402	2.074	55.857	57.723	38.517	51.568	241.643	263.596	1.083
	ElasticFace [ 23 ]	cumulative	23.646	40.996	17.973	18.768	5.867	8.173	1.630	2.457	59.054	60.379	39.146	51.597	205.884	277.024	1.042

TABLE II: Comparison of AUC-EDC and pAUC-EDC (lower is better) under the same experimental setting as Tab. I , but with CurricularFace loss instead of ArcFace.

Fig. 4: Error-versus-Discard Characteristic (EDC) curves for FNMR@FMR=

1e-3

of our proposed method CARPM-FIQA in comparison to SOTA. Results shown on eight benchmark datasets: LFW [ 14 ] , AgeDB-30 [ 15 ] , CFP-FP [ 16 ] , CALFW [ 17 ] , Adience [ 18 ] , CPLFW [ 19 ] , XQLFW [ 20 ] , and IJB-C [ 21 ] , using ArcFace [ 22 ] , ElasticFace [ 23 ] , MagFace [ 10 ] , and CurricularFace [ 24 ] FR models.

Configuration	$T$	$k$ @ $\rho{=}0.95$	$k$ @ $\rho{=}0.99$	first- $k$	last- $k$
ArcFace, cumulative, small	34	10	24	0.939	0.845
ArcFace, non-cumulative, small	34	11	24	0.938	0.857
CurricularFace, cumulative, small	34	11	24	0.935	0.854
CurricularFace, non-cumulative, small	34	11	24	0.934	0.869
ArcFace, cumulative, large	18	8	14	0.876	0.841
ArcFace, non-cumulative, large	18	8	14	0.883	0.840

TABLE III: Influence of the accumulation window: window length

k

(of

T

training epochs) at which the growing first-

k

window reaches Spearman rank correlation

\rho\geq 0.95

0.99

against the full-trajectory target

Q_{T}

, and Spearman correlation with

Q_{T}

of first-

k

vs. last-

k

windows at a matched relative window length (

k\approx 24\%

T

k{=}8

of 34 for the small protocol,

k{=}4

of 18 for the large).

Explore similar work

ATTN-FIQA: Interpretable Attention-based Face Image Quality Assessment with Vision Transformers
Apr 21, 2026Guray Ozgur, Tahar Chettaoui, Eduarda Caldeira +5Image Quality Assessment
Fraunhofer Institute for Computer Graphics Research IGD, Darmstadt, Germany · Department of Computer Science, TU Darmstadt, Germany
PreFIQs: Face Image Quality Is What Survives Pruning
May 13, 2026Jan Niklas Kolf, Guray Ozgur, Andrea Atzori +4Image Quality Assessment
FRAUNHOFER INSTITUTE FOR COMPUTER GRAPHICS RESEARCH IGD, GERMANY · TECHNICAL UNIVERSITY OF DARMSTADT, GERMANY · UNIVERSITY OF LJUBLJANA, SLOVENIA
Beyond Error-vs-Discard Characteristic: Toward Stable and Reliable Evaluation for Face Image Quality Assessment
Jul 23, 2026Bhavesh Wani, Žiga Babnik, Vitomir Štruc +1Image Quality Assessment
Johannes Gutenberg University Mainz, Germany · University of Ljubljana, Slovenia

Apr 21, 2026J/K move · Enter open · S save

ATTN-FIQA: Interpretable Attention-based Face Image Quality Assessment with Vision Transformers

Guray Ozgur, Tahar Chettaoui, Eduarda Caldeira, Jan Niklas Kolf+4

Fraunhofer Institute for Computer Graphics Research IGD, Darmstadt, Germany · Department of Computer Science, TU Darmstadt, Germany

Image Quality Assessment

Open paper AskPDF arXiv

Face Image Quality Assessment (FIQA) aims to assess the recognition utility of face samples and is essential for reliable face recognition (FR) systems. Existing approaches require computationally expensive procedures such as multiple forward passes, backpropagation, or additional training, and only recent work has focused on the use of Vision Transformers. Recent studies highlighted that these architectures inherently function as saliency learners with attention patterns naturally encoding spatial importance. This work proposes ATTN-FIQA, a novel training-free approach that investigates whether pre-softmax attention scores from pre-trained Vision Transformer-based face recognition models can serve as quality indicators. We hypothesize that attention magnitudes intrinsically encode quality: high-quality images with discriminative facial features enable strong query-key alignments producing focused, high-magnitude attention patterns, while degraded images generate diffuse, low-magnitude patterns. ATTN-FIQA extracts pre-softmax attention matrices from the final transformer block, aggregate multi-head attention information across all patches, and compute image-level quality scores through simple averaging, requiring only a single forward pass through pre-trained models without architectural modifications, backpropagation, or additional training. Through comprehensive evaluation across eight benchmark datasets and four FR models, this work demonstrates that attention-based quality scores effectively correlate with face image quality and provide spatial interpretability, revealing which facial regions contribute most to quality determination.

Abstract

Explore similar work

ATTN-FIQA: Interpretable Attention-based Face Image Quality Assessment with Vision Transformers

PreFIQs: Face Image Quality Is What Survives Pruning

Beyond Error-vs-Discard Characteristic: Toward Stable and Reliable Evaluation for Face Image Quality Assessment