Signature recognition application
- Neural Network
Conversion of Matlab signature recognition application to C# with EmguCV/OpenCV
Our developers have successfully completed the task of converting Matlab signature recognition application to C# (with EmguCV/OpenCV).
The project included transforming of an existing Matlab application into C# language. The deliverable consists of a dll library and user-friendly interface. An important requirement was the ability to use the code and the application stand-alone on machines without Matlab environment .
The original application leveraged a 3rd party C++ library that was missing along with some dependencies in the code. That added complexity to the whole project. Our developers have made a deep analysis of the library using special reverse engineering tools, restored dependencies and built a wrapper for the library on C# language.
Image processing functions were based on both available OpenCV functionality and also written on C# according to existing algorithms.
The final C# software has successfully passed QA on several versions of OS. Results generated by the resulting application and the original one match perfectly.
Virtual try-on tool for makeup products
The system consists of a face detection and segmentation model and an algorithm that allows recoloring objects without losing their original texture.
Online sign language interpreter
AI algorithm that converts video of a person using sign language into a text transcript
Workout helper app
Mobile app for the estimation of proper body positions during the workout.