UBIRIS

Departamento de Inform?ica da Universidade da Beira Interior

Contact:

SOCIA Lab. - Soft Computing and Image Analysis Group

Department of Computer Science, University of Beira Interior, 6201-001 Covilh\UTF{201E} Portugal

hugomcp@di.ubi.pt

lfbaa@di.ubi.pt

UBIRIS

Noisy Visible Wavelength Iris Image Databases

News:

29-03-2021: The UBIRIS data have been updated! We are releasing two novel versions of the original data.

1) For all the UBIPr samples, we now provide their segmentation masks (delimiting the "skin", "eyebrows", "sclera" and "iris"), given as a gray scale mask with the same name and ".png" format, with respect to the original sample. Some examples are given below:

2) We also provide a "Both Eyes" version of the periocular data, also with the corresponding segmentation masks (delimiting the "skin", "eyebrows", "sclera" and "iris"), in gray scale/".png" format. Examples are also given below:

Both data sets are available at the "UbiPr" section. They are freely available, under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International license.

01-04-2020: We are pleased to announce the availability of the first (as per April, 2020) fully annotated freely available data set for supporting the research about pedestrian 1) detection; 2) tracking; 3) re-identification and 4) search methods from aerial data.

As a tool to support the research on pedestrian detection, tracking, re-identification and search methods, the P-DESTRE is a multi-session dataset of videos of pedestrians in outdoor public environments, fully annotated at the frame level for:

1) ID. Each pedestrian has a unique identifier that is kept among the data acquisition sessions, which enables to use the dataset for pedestrian re-identification/search;

2) Bounding box. The relative position of each pedestrian in the scene is provided as a bounding box, for every frame of the dataset, which also enables to use the data for object detection/semantic segmentation purposes;

3) Soft biometrics. Each subject of the dataset is fully characterised using 16 labels: gender, age, height, body volume, ethnicity, hair color, hairstyle, beard, mustache, glasses, head accessories, action, accessories and clothing information (x3), which enables to use the dataset also for evaluating soft biometrics inference techniques.

The data set is available at [http://p-destre.di.ubi.pt/]

Within the biometrics context, the iris is commonly accepted as one of the most accurate biometric traits and has been successfully applied in such distinct domains as airport check-in or refugee control. However, for the sake of accuracy, present iris recognition systems require that subjects stand close (less than two meters) to the imaging camera and look for a period of about three seconds until the data is captured. This cooperative behavior is required to capture images with enough quality for the recognition task. However, it simultaneously restricts the range of domains where iris recognition can be applied, especially those where the subjects cooperation is not expectable (e.g., criminal/terrorist seek, missing children).

The main focus of the UBIRIS database is to minimize the requirement of user cooperation, i.e., the analysis and proposal of methods for the automatic recognition of individuals, using images of their iris captured at-a-distance and minimizing the required degree of cooperation from the users, probably even in the covert mode.

The UBIRIS database has two distinct versions:

- UBIRIS.v1 - This version of the database is composed of 1 877 images collected from 241 eyes during September, 2004 in two distinct sessions. It simulates less constrained imaging conditions. It is public and free available. In order to access the database, it must be sent an email to one of the authors asking for the password of the ?zip? file.

- UBIRIS.v2 - The second version of the UBIRIS database has over 11 000 images (and continuously growing) and more realistic noise factors. Images were actually captured at-a-distance and on-the-move.

- UBIPr - This is a version of the UBIRIS.v2 database, with images cropped covering wider portions of the ocular region. It is particularly suited for experiments on periocular recognition.

fghfghfghfgh

DI-UBI Bloco VI Rua Marqu? de ?vila e Bolama P- 6201-001 Covilh?PORTUGAL