Hello, my name is

YOGESH KUMAR BALJEET SINGH



A Deep Learning / AI enthusiast currently pursuing my Masters at the Saarland University



About me

I am Yogesh, a Master's student in Computer Science at Saarland University, and a part-time research assistant at K|Lens GmbH. I am currently in my final phase of studies. With a lot of motivation, I have focussed my course towards Artificial Intelligence and Data Science. I worked as a machine learning assistant for almost a year at August-Wilhelm Scheer Institut GmbH in Saarbrücken. I interned for 6-months as a Machine Learning Engineer at Bosch Center for Artificial Intelligence in Renningen. My focus is to dive deeper into the research of the state-of-the-art deep learning models, and hone my technical and research skills. I am currently pursuing my masters' thesis at the Robotics and Mechatronics institute of the German Aerospace Center (DLR) in the areas of Deep Learning and Computer Vision.

That said, I am really passionate about these topics and would like to work towards my career in the field of Deep Learning / AI and Software Development.

Interests

German Aerospace Center (DLR) - Masters' Thesis Student

Oberpfaffenhofen, Germany

January 2021 - Present

Working on the implementation of ‘Learning Joint Representations of a Robot using Motion’ using Siamese networks and Transformers.

K|Lens GmbH - Deep Learning Research Assistant

Saarbrücken, Germany

January 2021 - Present

Working on video depth estimation and video deflickering. Developed median-deflickering algorithm which calculates depth, optical flow and performs warping operation and finally median filtering.

German Aerospace Center (DLR) - Werkstudent

Oberpfaffenhofen, Germany

October 2020 - December 2020

Worked on the development of a dataset generation pipeline using BlenderProc. This pipeline will be soon published to the BlenderProc open-source repository.

German Aerospace Center (DLR) - Data Analytics Werkstudent

Braunschweig, Germany

June 2020 - August 2020

Pioneered the development of a new Data Analysis and Visualization tool based on simulations of Pilots and Air Traffic Control for performing debriefing using Django and deep learning.

Bosch Center for Artificial Intelligence - Data Science and Machine Learning Intern

Renningen, Germany

November 2019 - April 2020

Worked on the Web development and implementation of Deep Learning algorithms (Forecasting, Clustering, Classification and Detection) at the Bosch Centre for Artificial Intelligence.

August-Wilhelm Scheer Institut GmbH - Machine Learning Researcher

Saarbrücken, Germany

November 2018 - October 2019

Laid the groundwork for the architecture development and implementation of a demonstrator which employed LSTM model for online Gesture Recognition along with live connection to a sensor worn on hand using Flask.

Genesys - Associate Software Engineer

Chennai, India

October 2017 - July 2018

Worked on Genesys’s Product ’Composer’ (an Eclipse Plug-in) which is used to develop call-flow strategies.

Solarillion Foundation - Research Assitant

Chennai, India

April 2017 - June 2018

Worked on a machine learning based gesture recognition system and published the paper "A Generic Multi-modal Dynamic Gesture Recognition System using Machine Learning" in IEEE.

Zoho - Technical Staff Intern

Chennai, India

December 2016 - March 2017

Worked on the full-stack development of a survey feature in the company's product - 'Zoho Survey'

Education

Universität des Saarlandes - Masters' Student

Saarbrücken, Germany

October 2018 - Present

Working towards my masters' in Computer Science- focussed on Deep Learning/AI and Software Development

Anna University (Velammal Engineering College) - Bachelors' Student

Chennai, India

July 2013 - April 2017

Completed my Bachelors In Electronics and Communication Engineering

Projects

End-to-end Segmentation of Brain Tumor and Deployment

In this work, I develop a segmentation network using U-Net to segment the tumor from the input image. This pipeline is hosted on Heroku with continuous integration.

StackOverflow Survey 2019 & 2020 Analysis

In this work, I analyse the dataset of StackOverflow survey from years 2019-2020 to answer interesting questions using EDA

Covid-19 Diet

In this work, we used Machine Learning, to analyze the food consumption from countries all over the world to see whether there is a relationship between given country's food habit and their recovery rate

PhotoSharingApp-EnShare

Developed an Android app which allows context-based private photos sharing feature as a part of Software Engineering course

Story Generation

Worked on a CNN-RNN based model to generate style based stories from a single input image

Time Machine GAN

Worked on a GAN model (pix2pix) on before and after dataset (self-crawled dataset)-images consisting of old and new images of a same place but different times (years apart)

Accessible Price Tags

Developed an IOT system with an Android application to help visually impaired people in shopping by a haptic price tag touch, along with Speech output

Debunking Images

Developed 3 different algorithms and compared those in detection of Fake Images using CNNs

Hobby Projects

3D Path-Finder

In this work, I have developed 3D-Pathfinder, which is a fluid 3D, easy to understand and a new way to visualize pathfinding algorithms built on Three.js

CIFAR10 Classifier with Web Interface

Developed a web-app which uses a deep model trained on CIFAR10 to classify images

Publications

A Generic Multi-modal Dynamic Gesture Recognition System using Machine Learning

Published - IEEE Future of Information and Communication Conference (FICC - 2018)

Singapore

April 2018

Motion Controlled Face Recognition Enabled by Authentication Robot

Published - International Journal of Advanced Research in Basic Engineering Sciences and Technology (IJARBEST)

Chennai, India

May 2017

Cosmic Ray Energy Exploitation For Propulsion (C.R.E.E.P)

Published - National Conference on Large Scale Multi-Disciplinary Systems of National Significance

ISRO-Sriharikota

June 2016

A.T.O.M (Artificial Intelligence Talking Omnipresent Machine)

Published - International Journal of Innovative and Emerging Research in Engineering Volume 3, Special Issue-1, ICSTSD 2016

Malaysia

May 2016

Extra-curriculars - Hackathons

NASA SpaceApps Hacakthon 2020 - Project-L.L.O.C.U.S.T

This project was declared as one of the Global Winner of NASA SpaceApps Hacakthon 2020. Project L.L.O.C.U.S.T identifies and predicts the probability of locust invasion as well as forecast locust’s movements in vulnerable areas.

TexasInstruments Robotic Hackathon 2020

This project was awarded 3rd place. Worked on RSLK MAX robot to implement a new use case - Smart Parking Bot

ActInSpace Hackathon 2020

In this project we developed an easy-to-use holistic space catalog that helps operators to make space objects-related decisions faster, more efficient, and more precise.

ZooHackathon Europe 2020 - N.O.A.H 2 S.E.A

N.O.A.H 2 S.E.A is an easy to use social impact app that enables citizens to capture, record, identify products with respect to endangered species and report them

Slash Hackathon 2020 - PlanetZero

In this project, we developed an app which uses machine learning to predcit which kind of vehicle the user is using to estimate carbon emissions and provides options to balance them

Hacktoberfest 2020

During this event, I contributed to several open-source projects through the month of October 2020

Certifications

Coursera Deep Learning Specialization

AGI STK Certification - Level 1

Get in touch

I am always interested to work in the fields of Machine Learning / Deep learning / Software development. If you find my profile interesting, please feel free to reach out to me.