Uvais Karni

Master Thesis

Veoneer•  January 2021 - September 2021

Pedestrian Multiple Object Tracking Using Deep Learning:- Implemented two Multi-Object Tracking approaches Deep SORT and SORT-OH. Tested the viability of DNN based features in tracking. Determined that Re-Identification features and detection quality are critical to the performance of the Object Tracking approach.

Project Intern

Ericsson•  September 2020 - January 2021

Air quality prediction using machine learning:- Utilizing Federated Learning, developed and implemented a Timeseries prediction model using LSTMs to forecast multivariant air pollutants values in Stockholm. Performed data visualisation to comprehend the data, feature selection based on SHAP values and correlation, data imputation of missing values and outlier detection.

Data Analyst Intern

Resulticks•  September 2018 - June 2019

Responsible for data interpretation, data mining and data visualisation. Developed an app using R Shiny to generate essential dynamic plots frequently utilised by the organisation on data directly (made the repetitive task of data visualization a lot quicker).

Uppsala University

Computer Science, MS•  September 2019 - December 2021

Specialisation: Machine Learning, Image Analysis and Computer Vision.