# Leveraging Large-Scale Face Datasets for Deep Periocular Recognition via Ocular Cropping

arXiv: https://arxiv.org/abs/2510.26294 · HTML: https://mlsanity.com/paper/2510.26294/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 32 references that are in the map.
- First posted on arXiv 30 Oct 2025.

## Builds on (most cited first, 20 of 20)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size](https://mlsanity.com/paper/1602.07360/) (2016)
- [VGGFace2: A dataset for recognising faces across pose and age](https://mlsanity.com/paper/1710.08092/) (2017)
- [MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognition](https://mlsanity.com/paper/1607.08221/) (2016)
- [CNN Features off-the-shelf: an Astounding Baseline for Recognition](https://mlsanity.com/paper/1403.6382/) (2014)
- [WebFace260M: A Benchmark for Million-Scale Deep Face Recognition](https://mlsanity.com/paper/2204.10149/) (2022)
- [A survey of face recognition techniques under occlusion](https://mlsanity.com/paper/2006.11366/) (2020)
- [A Survey on Periocular Biometrics Research](https://mlsanity.com/paper/1810.03360/) (2018)
- [Periocular Recognition Using CNN Features Off-the-Shelf](https://mlsanity.com/paper/1809.06157/) (2018)
- [Ocular Recognition Databases and Competitions: A Survey](https://mlsanity.com/paper/1911.09646/) (2019)
- [On the Effect of Selfie Beautification Filters on Face Detection and Recognition](https://mlsanity.com/paper/2110.08934/) (2021)
- [Periocular Biometrics and its Relevance to Partially Masked Faces: A Survey](https://mlsanity.com/paper/2203.15203/) (2022)
- [Periocular Biometrics: A Modality for Unconstrained Scenarios](https://mlsanity.com/paper/2212.13792/) (2022)
- [Cross-Sensor Periocular Biometrics in a Global Pandemic: Comparative Benchmark and Novel Multialgorithmic Approach](https://mlsanity.com/paper/1902.08123/) (2019)
- [SqueezeFacePoseNet: Lightweight Face Verification Across Different Poses for Mobile Platforms](https://mlsanity.com/paper/2007.08566/) (2020)
- [A New Periocular Dataset Collected by Mobile Devices in Unconstrained Scenarios](https://mlsanity.com/paper/2011.12427/) (2020)
- [One-Shot Learning for Periocular Recognition: Exploring the Effect of Domain Adaptation and Data Bias on Deep Representations](https://mlsanity.com/paper/2307.05128/) (2023)
- [An Explainable Model-Agnostic Algorithm for CNN-based Biometrics Verification](https://mlsanity.com/paper/2307.13428/) (2023)
- [Combined CNN and ViT features off-the-shelf: Another astounding baseline for recognition](https://mlsanity.com/paper/2407.19472/) (2024)

## Abstract (arXiv)

We focus on ocular biometrics, specifically the periocular region (the area around the eye), which offers high discrimination and minimal acquisition constraints. We evaluate three Convolutional Neural Network architectures of varying depth and complexity to assess their effectiveness for periocular recognition. The networks are trained on 1,907,572 ocular crops extracted from the large-scale VGGFace2 database. This significantly contrasts with existing works, which typically rely on small-scale periocular datasets for training having only a few thousand images. Experiments are conducted with ocular images from VGGFace2-Pose, a subset of VGGFace2 containing in-the-wild face images, and the UFPR-Periocular database, which consists of selfies captured via mobile devices with user guidance on the screen. Due to the uncontrolled conditions of VGGFace2, the Equal Error Rates (EERs) obtained with ocular crops range from 9-15%, noticeably higher than the 3-6% EERs achieved using full-face images. In contrast, UFPR-Periocular yields significantly better performance (EERs of 1-2%), thanks to higher image quality and more consistent acquisition protocols. To the best of our knowledge, these are the lowest reported EERs on the UFPR dataset to date.