| Deep Photo Style Transfer

This project is a PyTorch implementation of the academic paper ‘Deep Photo Style Transfer’ by F. Luan et al. for photorealistic style transfers between the images.

| Ayurgenomics Viz Ml

In the Ayurveda system of medicine, individuals are classified into seven constitution types, “Prakriti”, for assessing disease susceptibility and drug responsiveness, which are assigned to an individual based on these phenotypes. We aim to develop machine learning algorithms for visualizing this heterogenous multi-dimensional phenomics and genomics data.

| Firewalls Using Prolog

The project was done in partial fulfillment of the course Logic in Computer Science during my under graduation at BITS Pilani.

| E Voting System Using Blockchain

The projects intends to build a proof of concept to make the election process in India more secure and reliable using Blockchain technology. This is also our submission for Codefundo 2019 organized by Microsoft.

| Ebdm

Implemented the research paper ‘A Unified Entropy-Based Distance Metric for Ordinal-and-Nominal-Attribute Data Clustering’ by Zhang et al. and built a python package for finding a common distance matrix for the ordinal and nominal data from any kind of questionnaire data, based on entropy measures. Tested the package on multiple datasets for robustness.

| Compiler Project

The project builds a compiler in C for a toy language called ERPLAG. The implementation includes 4 phases: lexical analysis, syntax analysis, semantic checks and code generation.

| Socket Programming

In this project, I worked on establishing a TCP connection and UDP connection between client and server nodes in absence/presence of intermediate(relay) nodes with out of order packet handling in c.

DeepBeats

Worked under Rintu Kutum, PhD Student, CSIR-IGIB and used OpenCV for creating a bounding box for the heart of Zebrafish in grayscale videos captured in a controlled environment. Analyzed variations in the segmented regions across the frames to measure the heartbeat of the Zebrafish.

Video AnalysisImage ProcessingOpenCVPython

Human Detection and Segmentation in Survillience Videos

The project does a comparative study of various state-of-the-art algorithms used in the broad domain of Computer Vision and compare them on the basis of their results on the COCO dataset for images and TownCentre and VIRAT dataset for pedestrian detection in videos.

Computer VisionHuman DetectionDeep Learning