Computer Science enthusiast, Machine Learning and AI explorer, Grace Hopper India Scholar, Open Source Contributer and a big travel freak!!

Detailed CV

About Me

Hi! I'm Samriddhi Jain, final year CSE undergraduate at Indian Institute of Technology Mandi. I'm interested in exploring new technologies in a hope of learning new technologies and applying them to solve real world problems. My areas of interest are Machine Learning, Artificial Intelligence, Logic Programming and Distributed Computing.

My other hobbies are sketching, travelling and dramatics.

Industrial Experience

I recently did a summer internship with Microsoft India Development Center, Hyderabad, where I worked with Business Continuity and Disaster Recovery (BCDR) team under Cloud and Enterprises group on the project Azure Backup Reporting. Everyday was a new challenge and it was a great learning opportunity working as a part of such a big community.

Research and Publications

Semantic Multinomial Representation for Scene Images Using Dynamic Kernel based SVMs.

(Accepted at SUNw at CVPR, 2016)

Guide: Dr. A D Dileep (IIT Mandi), Dr. Veena Thenkanidiyoor (NIT Goa)

Extracting and combining semantic pseudo concepts in an image with lower level image information for semantic multinomial representation and classification.

Object and activity based video summarization for vehicle mounted/egocentric cameras.

(ongoing final year technical project)

Guide: Dr. Renu M. Rameshan (IIT Mandi), Dr. Aditya Nigam (IIT Mandi)

Given a picture of fine object/scenario extract and summarize all the frames it is present in, that too in real time. Currently exploring the state of art techniques like densecap for coarse classification and deepmatching for fine grained matching.

Projects

Natural Language Processing for News-Tweets integration

Integrating Social Channels with Traditional News Media for holistic news browsing. Given a news headline extract the key entities, discover tweets which talk about these entities and perform a temporal analysis to integrate news and tweets. Attach real time tweets with different sections of news by matching the context. Analysing social behaviour using forced directed graphs, zipfian distribution and sentiment analysis.

Analytical and Qualitative assessment of SciDB datastore

Understanding the use of SciDb in multidimensional data storage and analytics in the scientific community. Study key differences between RDBMS and SciDB for various tasks and use MODIS(NASA) data in SciDB.

ColorIt Android Application

Colorblind Assist Application which allows users to capture an image and process it in such a way that the features are then easily differentiable by a colorblind person.

Digits Recognising GUI using supervised learning

Detects digits written on a canvas, uses CART and Random Forests classifiers for recognition. · Implemented using opencv 2.0, tkinter and sklearn library in Python.

Collaborative Real-time text editor

Text editing tool collaborative over multiple systems built on server client model using Express Node.js. Involves techniques and libraries like Operational Transformation, diff-match-patch library for data synchronisation.

Apache Spark Cluster computing framework

Understanding the key characteristics of Apache Spark framework like Resilient Distributed Datasets, fault tolerance, scalability that make them superior to their traditional counterparts like Hadoop, MapReduce. Implementing a Movie Recommender system in Apache Spark.

Other minor projects like bot for game ”Battle of Kings”, IIT Mandi Room Reservation System, 3D reconstruction and depth estimation from images, Object tracking in videos, High Altitude Ballooning Probe, Sudo-ku Solver.

Contact

If you have any suggestions or want to talk about something, I would love to hear from you, please feel free to contact.