Simultaneous Iris and Periocular Region Detection Using Coarse Annotations

  title={Simultaneous Iris and Periocular Region Detection Using Coarse Annotations},
  author={Diego Rafael Lucio and Rayson Laroca and Luiz A. Zanlorensi and Gladston J. P. Moreira and D. Menotti},
  journal={2019 32nd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)},
  • D. Lucio, Rayson Laroca, +2 authors D. Menotti
  • Published 31 July 2019
  • Computer Science
  • 2019 32nd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)
In this work, we propose to detect the iris and periocular regions simultaneously using coarse annotations and two well-known object detectors: YOLOv2 and Faster R-CNN. We believe coarse annotations can be used in recognition systems based on the iris and periocular regions, given the much smaller engineering effort required to manually annotate the training images. We manually made coarse annotations of the iris and periocular regions (≈122K images from the visible (VIS) spectrum and ≈38K… 

Figures and Tables from this paper

Deep Representations for Cross-spectral Ocular Biometrics
This study designs and extensively evaluates cross-spectral ocular verification methods using well known deep learning representations based on the iris and periocular regions, and fine-tune convolutional neural network models, originally trained for face recognition.
UFPR-Periocular: A Periocular Dataset Collected by Mobile Devices in Unconstrained Scenarios
A new periocular dataset containing samples from 1,122 subjects, acquired in 3 sessions by 196 different mobile devices is presented, and an extensive benchmark with several Convolutional Neural Network architectures and models that have been employed in state-of-the-art approaches based on Multi-class Classification, Multitask Learning, Pairwise Filters Network, and Siamese Network is performed.
CNN Hyperparameter tuning applied to Iris Liveness Detection
A hyperparameter tuning of the CASIA algorithm, submitted by the Chinese Academy of Sciences to the third competition of Iris Liveness Detection, in 2017 is proposed and promoted an overall improvement.
Generalizable deep features for ocular biometrics
An efficient deep learning-based feature extraction pipeline for learning the aforementioned generalizable features for ocular recognition using a relatively small Convolutional Neural Network (CNN) based feature extraction model along with a region of interest (ROI) detector and data augmenter.
Iris Recognition Development Techniques: A Comprehensive Review
The researchers present the techniques used in different phases of the recognition system of the iris image and explains the two approaches of iris recognition which are the traditional approach and the deep learning approach.
International Journal of Recent Technology and Engineering (IJRTE)
46 Retrieval Number: F10100386S20/2020©BEIESP DOI:10.35940/ijrte.F1010.0386S20 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Abstract: Real time simulators play a major role
Pupilometria Dinâmica para Avaliação do Reflexo Foto Motor na Detecção da Neuropatia Autonômica Diabética e Relação Glicêmica
  • B. L. Bonafini, G. L. F. Ronque, Lucas F. de Oliveira
  • Anais Estendidos do XXI Simpósio Brasileiro de Computação Aplicada à Saúde (SBCAS Estendido 2021)
  • 2021
A Neuropatia Autonômica é uma das complicações da diabetes por ocasionar danos ao sistema nervoso e uma forma de acessá-lo é por meio da análise da reação pupilar. Este estudo teve como objetivo
Ocular Recognition Databases and Competitions: A Survey
A survey of the databases created for ocular recognition, detailing their protocols and how their images were acquired and some relevant works applying deep learning techniques to ocular Recognition point out new challenges and future directions.


A Benchmark for Iris Location and a Deep Learning Detector Evaluation
Experimental results showed that the deep learning based detector outperforms the other ones in terms of accuracy and runtime (GPUs version) and should be chosen whenever possible.
Joint Iris Segmentation and Localization Using Deep Multi-task Learning Framework
This paper proposes a deep multi-task learning framework, named as IrisParseNet, to exploit the inherent correlations between pupil, iris and sclera to boost up the performance of iris segmentation and localization in a unified model.
The Impact of Preprocessing on Deep Representations for Iris Recognition on Unconstrained Environments
The use of deep representations, more specifically, architectures based on VGG and ResNet-50 networks, are proposed for dealing with the images using (and not) iris segmentation and normalization, and the approach using non-normalized and only circle-delimited iris images reaches a new state of the art in the official protocol of the NICE.
Fusing iris and periocular information for cross-sensor recognition
The application of well-known iris and periocular recognition strategies based on classical encoding and matching techniques are proposed, as well as demonstrating how they can be combined to overcome the issues associated with mobile environments.
Human identification from at-a-distance images by simultaneously exploiting iris and periocular features
  • Chun-Wei Tan, Ajay Kumar
  • Computer Science
    Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012)
  • 2012
A completely automated joint iris and periocular recognition approach from the face images acquired at-a-distance using Leung-Mallik filters to achieve 8.1% improvement in recognition accuracy over the best performing approach among SIFT, LBP and HoG presented in the literature.
Robust Iris Segmentation Based on Fully Convolutional Networks and Generative Adversarial Networks
Two approaches for robust iris segmentation based on Fully Convolutional Networks (FCNs) and Generative Adversarial Networks (GANs) are described and achieved promising results in all evaluated datasets.
Periocular Biometrics in the Visible Spectrum
The feasibility of using the periocular region as a biometric trait is studied, including the effectiveness of incorporating the eyebrows, and use of side information (left or right) in matching.
Segmenting the periocular region using a hierarchical graphical model fed by texture / shape information and geometrical constraints
An integrated algorithm for labelling the periocular region is described, that uses a unique model to discriminate between seven components in a single-shot: iris, sclera, eyelashes, eyebrows, hair, skin and glasses.
A novel eyebrow segmentation and eyebrow shape-based identification
A fully automatic, robust and fast graph-cut based eyebrow segmentation technique to extract the eyebrow shape from a given face image and an eyebrow shape-based identification system for periocular face recognition are proposed.
A Robust Real-Time Automatic License Plate Recognition Based on the YOLO Detector
A robust and efficient ALPR system based on the state-of-the-art YOLO object detector is presented, employing simple data augmentation tricks such as inverted License Plates and flipped characters for character segmentation and recognition.