cs.CVAug 10, 2026

Periocular Soft Biometrics: A Survey and Applications to Multimedia Forensics and Disinformation Detection

Authors: Fernando Alonso-Fernandez, Kevin Hernandez-Diaz, Josef Bigun

Organizations: School of Information Technology (ITE), Halmstad University, Sweden

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Soft-biometric attributes such as gender, age, and ethnicity provide valuable ancillary evidence when full identity recognition is not feasible, supporting applications in forensic investigation, identity verification, surveillance, or detection of synthetic and manipulated media. Among biometric modalities, the periocular region is a robust source of soft-biometric cues, as it often remains visible when other parts of the face are occluded, a frequent condition in forensic evidence and surveillance footage, and can be captured across a wide range of acquisition conditions. In this paper, we provide a survey of demographic attribute estimation from periocular images, covering publicly available datasets, methodological trends from handcrafted descriptors to deep learning architectures, and the state of the art in gender, age, and ethnicity prediction. We discuss use cases relevant to multimedia forensics and disinformation-detection applications, including demographic filtering in surveillance footage, age verification, and the detection of demographic inconsistencies in synthetic data. We also highlight open challenges, including dataset bias, cross-domain generalisation, fairness, ethical aspects, and the lack of forensic-oriented benchmarks.

Figures & tables

Figure 1: From left to right: ii ) anatomical parts of the periocular region (the periocular modality uses the full visible area around the eye, not just the iris region, image from [ 1 ] ); iiii ) demographic indicators; iiiiii ) limitations and open challenges in periocular soft biometrics research; and iviv ) applications of periocular soft biometrics to forensic and disinformation detection applications.
Figure 2: Representative samples from the three dataset categories (face/iris/periocular) used in periocular demographic research.
       Best accuracy
Ref.YearDatabaseTypeImagesSubj.SpectrumGenderAgeEthnicity
[ 39 ]1998ARface+4000126VIS85.61%--
[ 51 ]2000FERETface141261199VIS90%-98.5%
[ 50 ]2005FRGCface36818741VIS97.5%-98.7%
[ 57 ]2006MORPH-IIface55k13kVIS-3.63 yr-
[ 26 ]2007LFWface132335749VIS93.4%60.2%82.9%
Table 1: Publicly available datasets used in periocular demographic research. VIS=Visible spectrum. NIR=Near-infrared spectrum. V+N=VIS+NIR. (V)=Videos. Best accuracy columns give the top reported value in the literature (Tables 2 and 3 ).
PERIOCULAR GENDER ESTIMATION
Ref.YearMethodDatabaseSpec-trumEyesAccu-racy
[ 34 ]2012ICA + NNFERETVISBoth90%
[ 54 ]2017(LBP, LTP, LPQ, BSIF, HOG) + (SVM, MLP)VISOBVISOne90.2%
[ 62 ]2017Deep Class-EncoderND-Iris-0405NIROne82.53%
[ 67 ]2017(Intensity, Texture, Shape) + RFCross-EyedV+NOne90%
[ 7 ]2019compass LBP + SVMAdienceVISOne Both84.06% 83.27%
Table 2: Selected results for periocular gender estimation (best results of each database). VIS=Visible spectrum. NIR=Near-infrared spectrum. V+N=VIS+NIR.
PERIOCULAR AGE ESTIMATION
Ref.YearMethodDatabaseSpec-trumClassesEyesAccuracy
[ 73 ]201423 sub-CNNsMORPH-IIVIS62Face parts3.63 yr (MAE)
[ 16 ]20194 sub-CNNsAdienceVIS8Face parts51.03% (group) 83.41% (1-off)
[ 3 ]2020CNN + SVMLFWVIS3One Both60.2% (group) 60.0% (group)
PERIOCULAR ETHNICITY ESTIMATION
Ref.YearMethodDatabaseSpec-trumClassesEyesAccuracy
Table 3: Selected results for periocular age and ethnicity estimation (best results of each database). VIS=Visible spectrum. NIR=Near-infrared spectrum. MAE=Mean Absolute Error in years. AS=Asian, NAS=no-Asian, ME=Middle-East, NME=no-Middle-East, CA=Caucasian, CH=Chinese, IN=Indian, WH=White, BL=Black, OT=Other. The name of the ethnic classes are as defined by the authors of the respective papers mentioned in the table.
More in this topic

Biometric Identification

Biometric identification matches or verifies people and animals using distinctive physical, physiological, or behavioral signals.

See all 128 papers

Explore similar work

CardsList
May 15, 2026

Attention-Aware Transformer-Based Aggregation Network for Video Periocular Recognition

Video periocular recognition is the task of recognizing an individual's identity based on the region around an individual's eyes. The periocular area is one of the most discriminative regions of the human face, making it suitable for recognition tasks. Its use as a biometric modality has emerged as an alternative, especially in surveillance scenarios where conventional biometric traits such as face or iris recognition become unfeasible due to unconstrained acquisition conditions. This paper proposes an attention-aware approach for video-based periocular recognition in surveillance environments. The framework consists of two main modules: feature embedding and aggregation. The feature embedding module is a deep convolutional neural network that maps periocular data to feature vectors. The aggregation module is an encoder-only transformer that adaptively learns to aggregate frame-level features into a single video representation and a feature vector for the still reference image. Experiments on the publicly available COX Face dataset show the robustness of the proposed method, consistently outperforming naive aggregation schemes. In the best scenario, the approach achieves 99.8%99.8\% of TPR@1e−11e^{-1} and 96.6%96.6\% of Rank-5.

Luiz G F Carreira, Breno A Mariano, Victor H C de Melo +2

Department of Computer Science, Federal University of Minas Gerais, Belo Horizonte, Brazil · Department of Informatics, Federal University of Paran´a, Curitiba, Brazil

Biometric IdentificationFace Recognition
AskPDF
Jun 3, 2026

Toward Calibrated, Fair, and accurate Deepfake Detection

Deepfake detectors show large performance gaps across demographic groups. Existing fairness approaches require demographic labels, retraining, or sacrifice accuracy. We introduce Face-Fairness (FF), a plug-and-play framework for bias mitigation. Our primary contribution, Face-Feature Tuning (FFT), is the first demographic label-free fairness method demonstrated for deepfake detection: a lightweight calibrator that performs a logit remapping conditioned on frozen face embeddings. We complement FFT with two variants: FF-Max, which maximizes worst-group accuracy when demographics are available, and FF-Discover, which does the same with embedding-discovered groups. Across in-domain and cross-dataset test settings, FF consistently reduces FPR/TPR gaps and improves minimum group accuracy while maintaining (often improving) overall accuracy. The approach is detector-agnostic, adds negligible runtime overhead, and requires no access to identity attributes.

Ryan Brown, Chris Russell

University of Oxford

Algorithmic FairnessDeepfake DetectionModel Calibration
AskPDF
May 17, 2026

Position: Age Estimation Models Do Not Process Biometric Data

When a neural network estimates someone's age from a photograph, does it process biometric data? The answer depends on whether identity-discriminative representations arise within the network during inference, a question that may seem trivial to ML researchers but triggers consent requirements under GDPR, statutory damages under BIPA, or high-risk AI classification under the EU AI Act. Yet no regulatory guidance addresses it. This position paper provides empirical evidence: 14 models evaluated across 3 face verification benchmarks show age estimators fall orders of magnitude short of identification thresholds. Age estimation models cannot identify individuals. We call on researchers to provide transparency about what systems store and can do, and on regulators to distinguish transient processing from template storage.

Nikita Marshalkin

Sumsub GmbH, Berlin, Germany.

Biometric IdentificationAI AccountabilityRepresentation Identifiability
AskPDF