Madi Yussupov

Computer Vision Engineer

Sergek

Sergek is an intelligent video control, analysis and forecasting system that detects violations on the roads and streets of a city.

Improves vehicle tracking, with twice as many violations detected, and develops metrics for evaluating multi-vehicle tracking algorithms.

• Improving old car tracking algorithm. Twice more violations detected. • Developing a metrics for non-labelled video dataset to evaluate multivehicle tracking algorithms.

Computer Vision Engineer

Halyk Bank

Halyk Bank is a commercial savings bank in Kazakhstan that also has branches in Kyrgyzstan, Georgia, Russia, Tajikistan and Uzbekistan.

Built document OCR and information-extraction systems, a denoising segmentation model that reduced extraction errors by 50%, and signature and seal detection with 85% mean average precision.

• I developed a system to recognize text from scanned documents and extract target information to save man-hour. The system was REST-service written on Flask and Celery. Covered with tests. • Made a segmentation model that cleans scanned documents from noise. It decreased error rate of target values extraction from documents by 50%. • The automatic extraction of information from image of remote desktop connection. It helped to save man-hour and build automatic system.

Computer Vision Engineer

ReliveResearch

The company is developing projects related to deep learning and research that could be used in industrial field.

Worked on thermal-camera body-temperature measurement, person detection and alert management, and medical image classification.

• I developed a system of measuring person body temperature via with thermal imaging camera. It was built for COVID-19 pandemic preventive measurement. • Developing an Person Detection Algorithm for automatic security alert. • Developing an Flask for managing alerts from analysed videos. • Medical Image Classification.

Computer Vision Engineer

AV-software

It is IT company that is developing products in several directions. Including Machine Learning.

Contributed to face recognition and spoofing detection, and developed video analytics for counting customers and estimating their age and gender.

• I worked in team that developed Face Recognition system that tracked people entering and exiting time. Main responsibility of mine was to research and implement papers. • I developed Customers analyzing system. Via video analytical algorithm customers were counted, and described by age, and gender. It helped to understand the customers image. • I tested and implemented Face spoofing detection system. To secure the face recognition system from fraud attacks there was built algorithm to detect such matters.

Bachelor, Computer Systems and Software

Suleyman Demirel University