Désolé, cet article est seulement disponible en anglais américain. Pour le confort de l’utilisateur, le contenu est affiché ci-dessous dans une autre langue. Vous pouvez cliquer le lien pour changer de langue active.

Smart Engines is a corporate-academic institution focused on AI-related research. Our team is engaged in solving research problems related to document recognition, computational imaging and tomography, and consists of more than 60 scientists, including two Doctors of Science and 16 PhDs.

Smart Engines scientific activities were started by one of now-founders of the company, back in the days of the USSR. His research has resulted in developing the first program that could beat the world champions in chess.

Smart Engines core developers are faculty members of The Moscow Institute of Physics and Technology, professors at the Department of Cognitive Technologies. Their research papers are being published by the highest-rated scientific journals, and they are regular participants at the leading international conferences, such as ICDAR, ICIP, ICMV, etc.

Recently, in 2017, Smart Engines team won the international Document Image Binarization Competition, held within ICDAR — The International Conference on Document Analysis and Recognition, in Kyoto, Japan. A few years earlier we became the 3rd place winners at the global programming contest on smartphone document capture (ICDAR 2015, Nancy, France).

Career Opportunities

For internships and job opportunities with Smart Engines, please contact us via job@smartengines.ru.

MOSCOW INSTITUTE OF PHYSICS AND TECHNOLOGY

Smart Engines team is the core faculty of the Department of Cognitive Technologies at The Moscow Institute of Physics and Technology. The department offers students « Applied Mathematics and Computer Science » and « Applied Mathematics and Physics » bachelor and masters programs.

Taught courses:

• Effecient data structures and algorithms, their construction and analysis
• Modern programming languages ​​and platforms
• Information Security
• Modeling of Wheeled Robots
• Machine Learning and Neural Networks
• Industrial Recognition Systems
• Intelligent Information Systems
• Combinatorial Optimization Algorithms
• Speech synthesis and recognition technology
• Processing and analysis of images and video streams
• Optimization of calculations on modern processor architectures
• Technical vision and 3D scene recognition
• IT project management.

The Department of Cognitive Technologies is headed by Smart Engines’ Director of Science, Corresponding Member of the Russian American of Sciences, Professor Vladimir Arlazarov L.

RESEARCH AREAS

Document Recognition Technologies

Uncontrolled capture conditions and unknown parameters of the capture equipment are the main challenges in recognition of documents in the video stream and photos. When recognizing directly on a mobile device, the computational complexity of the methods used, as well as the size of the downloaded data, including parameters of the artificial neural network, are of crucial importance. The relevance of the data entered into the system is generally highly dependent on the user, which requires high fault tolerance of the algorithms that we create.

Computational Imaging and Tomography

Machine vision, being one of the most powerful non-destructive testing methods in the optical range, is still limited to studying only the surface of objects. To “look inside” and explore the three-dimensional internal structures of objects (required in medicine, for industrial diagnostics, in scientific laboratories), the use of computed tomography becomes necessary.

In the field of tomography, our team works on the scientific developments and practical introduction of the software, which solves the following tasks:

• calibration and alignment of new generation tomographs;
• optimized (customized) reconstruction of images from data collected in difficult conditions (ultra-low doses, tomosynthesis, the presence of highly absorbing inclusions in the object, etc.);
• computational visualization with automatic processing and semantic analysis of results.

INTERNATIONAL CONFERENCES

The 15th International Conference on Document Analysis and Recognition (ICDAR 2019)
September 20-25, 2019, Sydney, Australia
Proceed to ICDAR 2019 website

The 12th International Conference on Machine Vision (ICMV 2019)
November 16-18, 2019, Amsterdam, The Netherlands
The CEO of Smart Engines, Vladimir Arlazarov, Ph.D., is the Head of the Camera Based and Mobile Recognition scientific section at ICMV 2019.
The Technical Director of Smart Engines, Dmitry Nikolaev, Ph.D, is the chairman of the Advanced Imaging and Tomography scientific section at ICMV 2019.
Proceed to the ICMV 2019 website

MIDV-500 DOCUMENTS DATASET

The comprehensive study of the issues related to document recognition on mobile devices using machine learning methods is not possible with currently available public datasets, due to insufficiency of the latter. Existing datasets can be useful for certain tasks associated with processing document images on mobile devices, however more specialized datasets are required to create and test technologies for recognizing identity documents.
Smart Engines introduces the Mobile Identity Document Video (MIDV-500) video dataset, which consists of 500 video clips for 50 different types of identity documents. Since identity documents contain personal data, all source images of documents used in the MIDV-500 are either publicly available or their distribution does not infringe copyright.
Page with description and a download link for MIDV-500 dataset https://arxiv.org/abs/1807.05786

Return to Home Page





For any questions, project proposal or bespoke solution request,
please Reach us by phone +7 (495) 649-82-60
or fill the form below and we will get back to you.

Smart Engines garantit que les informations fournies ne seront pas rendues publiques et ne seront utilisées qu’en interne à des fins d’évaluation.

Warning before submitting your request:

Smart Engines is fully committed to provide an answer within 2 working days. However, it is your responsibility that your IT infrastructure does not block our reply or redirected it into your spams. If you haven’t received any answer from us within 2 working days, please check your spams or re-submit your request and write down your phone number (+ country code / number) in the message section or simply call us.