Skip to main content
Till KTH:s startsida Till KTH:s startsida

Abubakr Karali

Profile picture of Abubakr Karali




About me

I am interested in applied large scale computer vision, natural language processing and machine learning. For years, I dealt with a wide range of projects while working as research engineer at Microsoft Research, Data Scientist VionLabs AB and Computer Vision Lead at Research institute of Sweden (RISE), respectively. These projects includes 2D object detection and tracking, object recognition and scene classification, stereo imaging, camera calibration, 3D reconstruction, multi cameras systems interface and calibration, human emotion understating, face tracking, fashion analysis and human action recognition. I employed in this research the state of the arts machine learning techniques in supervised, self-supervised, semi supervised learning, active learning, multi-instance multi-label learning and multiple classifier systems. I also worked in graphics in medical data visualization, natural language processing and GPU computing. The state of the art techniques employed in these projects resulted into 9 publications and one patent.

Profile picture of Abubakr Karali