Disculpa, pero esta entrada está disponible sólo en inglés estadounidense. For the sake of viewer convenience, the content is shown below in the alternative language. You may click the link to switch the active language.

The R&D team of Smart Engines presented the brand new approaches to the reduction of carbon emission. Smart Engines committed to push forward the computer vision industry to become more ecological by developing low energy consumption OCR solutions that benefit the final users and for better preserving the resources of our planet.

Our research team took part in 15th international conference on document analysis and recognition (ICDAR) in Sydney, Australia. For almost 30 years, ICDAR has become a successful and a flagship scientific conference series, which gather international community of experts, researchers, scientists and practitioners in document analysis. In this conference fundamental and applied subjects are debated. It was a tremendous opportunity for our team of scientists to get in touch with the world of artificial intelligence, computer vision and software development leaders. Also we were able to demonstrate our latest scientific research results in the fields of computer vision.

We presented our recent scientific papers in ICDAR on the matter of optimal stopping strategies for text recognition in a video stream as an application of a monotone sequential decision model. Stopping the text field recognition process in a video stream is a novel problem, particularly relevant to real-time mobile document recognition systems. We provided decision-theoretic framework for this problem, and explored similarities with existing stopping rule problems.

Also our R&D team presented two reports during the poster session on ID analysis and recognition. In the first one, we presented our approach to solve problems related to simultaneous document type recognition and projective distortion parameters estimation for the images of ID documents. In the second one, we introduced a novel neural network architecture based on Fast Hough Transform layer. The layer of this type allows our neural network to accumulate features from linear areas across the entire image instead of local areas. Therefore, we have demonstrated its potential by solving the problem of vanishing points detection in the images of documents. Such problem occurs when dealing with camera shots of the documents in uncontrolled conditions. In this case, the document image can suffer several specific distortions including projective transformation.

Our company always pays great attention to develop the fundamentals of computer vision and AI implications in character recognition. Our research allowed us to create and improve one of the most powerful and precise recognition technologies for the world markets. The Smart Engines R&D team strives to be the quintessence of the computer vision industry pushing OCR technologies as far as possible. By doing so, the team is constantly publishing and disclosing the results of their technological advances in scientific journals, and when attending international scientific conferences.

 





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 guarantees that the provided information will not be made public and will be used only internally for evaluation purpose.

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.