► This website was developed keeping cybersecurity in mind. It helps you to geolocate any queried ip address across the world. The live preview for the website is here

► This is a progressive web application which can be easily installed on any device be it linux, windows, android or mac by just a click from browser itself. It uses openweather api to fetch the live weather of the queried location. The live preview for the website is here

► I Developed a ReactJS based web application that uses an api to fetch the present statistics and data related to covid 19 such as Infected, Recovered and Deaths. The live preview for the website is here

► Our team worked on a project to make the complex process of counting vehicle density manually easier by automating it using camera feeds. It involves use of computer vision and deep learning.

► As a part of Hackathon Challange, Our team developed a IoT based smart energy meter along with an entire online billing dashboard and an android application that lets you monitor your usage in realtime. It was backed by AI powered bill prediction system which provides guide to save energy and helps in energy saving.

► Our team worked on development of a smart watch with GPS integration and health monitoring systems such as body temprature and pulse rate. The entire data could be monitored through a single dashboard. In case of no GPS coverage It used AI based algorithms to predict accurate location.

► Rock paper scissor is a two player game. It works by recognising your hand posture through webcam feed and moves based on your past move history . The game has been trained to win almost 70% of the time.

► Neural style transfer is an optimization technique used to take two images—a content image and a style reference image (such as an artwork by a famous painter)—and blend them together so the output image looks like the content image, but “painted” in the style of the style reference image. You can try it online here

► AI based image recoloring of black and white images uses deep neural networks trained over a large sample of images in LAB color space with L channel as input and AB channel as output. Here is a video demonstrating recoloring of an Iconic Hindi Bollywood B&W song. This is an implementation of research paper titled 'Colorful Image Colorization' authored by Richard Zhang, Phillip Isola and Alexei Efros. The used model was trained over 1.5 Million images. You can check out the recoloured video here Also you can try it online over here

► Developed a smart city street light monitoring system. My team developed a dimmer module that could tune the intensity of light based upon the traffic detected. This enable for energy saving by dimming the lights when no traffic is detected. We also developed a smart system integrated with a flask backed web app that would allow the control station to monitor each and every light. In case of a failure in an of the street lights, it is automatically informed to the web app to allow easy and quick maintenance.