Trainings and Internships

Computational Cognitive Science, Universität Stuttgart

(May 2023 - Present) - Graduate Student Assistant

• Title: LSLAutoBIDS
• Supervisor: Prof. Benedikt Ehinger
• Project Title: LSLAutoBIDS: A tool for automatic conversion of EEG data to BIDS format
• Project Overview: Worked on the development of a python package to automate the conversion of raw EEG data to the BIDS format.
• The project is publicly available here.

Institut für Automatisierungstechnik und Softwaresysteme, Universität Stuttgart

(Feb 2023 - May 2023) - Graduate Student Assistant

• Supervisor: Dominik Braun
• Project Title: Frequent Subgraph mining on Neo4j database fir the Arena 2030 industrial setup
• Project Overview: Worked with Neo4j Graph Database system and Python for running frequent subgraph pattern mining algorithms for an Industrial graph database setup. Algorithms like gSpan and other community detection algorithms were explored to find patterns and generate templates for the dump file, using the neo4j plugins.

Community Climate Center, Azim Premji University

(May 2022 - Oct 2022) - Research Intern

• Supervisor: Prof. Santonu Goswami
• Project Title: Deep learning based flood mapping: A Case Study from a flood-vulnerable state in the Northeastern region of India
• Project Overview :In this work we created a deep learning-based flood model that can predict the flood vulnerability of a region based on various spatial inputs, like slope, elevation, historical flood data, etc. These models make use of convolutional layers which are able to extract the spatial information based on the inputs we provide. Unlike the hydrodynamic models, our deep learning-based approach, which uses the ResNet architecture as the backbone, attempts to reduce the inference time of flood prediction for a given set of inputs.
• Poster Presentation : This project was presented as a poster in the American Geophysical Union (AGU) Fall meeting 2022. Poster

• Supervisor: Prof. Santonu Goswami
• Project Title: Mapping Biogeophysical Parameters for Indian Cities Using Long-Term Satellite Data and Geospatial Technologies

• Project Overview : In this work, we have carried out an analysis to characterize the long-term changes in biophysical parameters such as vegetation, water bodies and built-up area for any specified target city or region, given the geographical boundaries. The dataset was obtained from the satellite data of Landsat 7. More about the project is available here.The final goal of the project aimed at building an automated pipeline which can now process long-term remotely sensed data and plot various biophysical indices eventually making the process of preliminary investigations of biophysical parameters fast by manifolds as well as efficient.

• Poster Presentation : This project was presented as a poster in the American Geophysical Union (AGU) Fall meeting 2022. Link to the poster : Poster

• Supervisor: Prof Manas Kamal Bhuyan
• Co Supervisor: H Pallav Jyoti Dutta
• Project Title: Analysis and Implementation of Optimized Algorithms for 2D Hand Pose Estimation.
• Project Overview: Experimented and Analysed various 2D keypoint localization deep learning architectures that are seminal or state of the art , and propose optimizations for the same. Self Attention based Convolutional Neural Networks were used as the based model to the accurate keypoints.(Tensorflow,Keras)

• School Overview: A month long summer school on Artificial Intelligence for a group of selected students and researchers of all levels to introduce the various domains of AI and their advancements.
• Link to notes

• Supervisor: Prof Nutan Kumari Singh
• Subjects: Engineering Physics, Computer Programming with C
• Overview: Responsible for conducting online classes to help out the freshman year undergraduates.

• Supervisor: Prof Layi Alatise
• Co Supervisor: Jose Ortiz Gonzalez
• Title: Development of a test setup for the evaluation of SiC Power Mosfets.
• Project Overview: Developed a lapview model to take current and temperature measurements via the temperature and voltage sensors by NI Instruments and analyse its characteristics in SiC Power Mosfets.

• Supervisor: Prof. Rhittwikraj Moudgollya
• Title: Image Processing based Object Tracking Turret
• Project Overview: Developed two prototypes for object detection for a surveillance task experimenting on various image processing and deep learning algorithms in an embedded steup. Raspberry PI 3B+ and NVIDIA Jetson Nano were used for experimentation and analysis.
• Link to Paper

• Supervisor: Prof Rakesh Matam
• Project Title: Engagement Learning using Deep Learning
• Project Overview: Developed a model to track the attention span of students in a classroom environment.(Tensorflow,Keras)

• Department: Instrumentation Engineering Department, IOCL
• Overview: Understanding the operations of the various instruments in the plant accross the process unit. The sub departments covered are Workshop, MSQ-CDU UNit, DCU Unit and Thermal Power Station
• Report : Link

• Department: Instrumentation Engineering Department, AEGCL
• Overview: Understanding the working of a grid station with various instruments and parts and its control procedure.
• Report : Link