MIDV-500: a dataset for identity document analysis and recognition on mobile devices in video stream

V.V Arlazarov,K Bulatov,V.L Arlazarov,T Chernov

doi:10.18287/2412-6179-2019-43-5-818-824

V.V Arlazarov, K Bulatov + Show 2 more

Open Access

https://doi.org/10.18287/2412-6179-2019-43-5-818-824

Copy DOI

Journal: Computer Optics	Publication Date: Oct 1, 2019
Citations: 75	License type: cc-by

Affiliation: Moscow Institute of Physics and Technology

Abstract

A lot of research has been devoted to identity documents analysis and recognition on mobile devices. However, no publicly available datasets designed for this particular problem currently exist. There are a few datasets which are useful for associated subtasks but in order to facilitate a more comprehensive scientific and technical approach to identity document recognition more specialized datasets are required. In this paper we present a Mobile Identity Document Video dataset (MIDV-500) consisting of 500 video clips for 50 different identity document types with ground truth which allows to perform research in a wide scope of document analysis problems. The paper presents characteristics of the dataset and evaluation results for existing methods of face detection, text line recognition, and document fields data extraction. Since an important feature of identity documents is their sensitiveness as they contain personal data, all source document images used in MIDV-500 are either in public domain or distributed under public copyright licenses. The main goal of this paper is to present a dataset. However, in addition and as a baseline, we present evaluation results for existing methods for face detection, text line recognition, and document data extraction, using the presented dataset.

Highlights

Smart phones and mobile devices have become the de-facto way of receiving various government and commercial services, including but not limited to e-government, fintech, banking and sharing economy [1]
Some classes of publicly available datasets may be useful for researching common subtasks of identity document analysis and recognition
In this paper we present a Mobile Identity Document Video dataset (MIDV-500), which in contrast to other relevant publicly available datasets can be used to develop, demonstrate and benchmark a coherent processing pipeline of identity document analysis and recognition in its modern applications and use cases

Summary

Introduction

Smart phones and mobile devices have become the de-facto way of receiving various government and commercial services, including but not limited to e-government, fintech, banking and sharing economy [1]. Many organizations involved in these areas decide to utilize identity document analysis systems in order to improve data input processes The goals of such systems are: to perform document data recognition and extraction, to prevent identity fraud by detecting document forgery or by checking whether the document is genuine and real, and others. Some classes of publicly available datasets may be useful for researching common subtasks of identity document analysis and recognition. In this paper we present a Mobile Identity Document Video dataset (MIDV-500), which in contrast to other relevant publicly available datasets can be used to develop, demonstrate and benchmark a coherent processing pipeline of identity document analysis and recognition in its modern applications and use cases. The dataset is available for download at ftp://smartengines.com/midv-500/

Problem statement and use case Identity document properties

Dataset structure

Experimental baselines

Findings

Conclusion