An app powered by a very small deep learning model (0.5mb) trained using contrastive supervised learning and sharpness aware minimization on Plant Village Dataset to detect plant leaf disease.

Tags: Contrastive Learning, Sharpness Aware Minimization, App, Model Optimization

While drunk or drowsy people can’t react to stimuli efficiently in the environment, thus we intend to check for a verbal response from the driver upon detecting anomalous driving patterns. IMU tracker upon detecting frequent changes in acceleration and sharp turns triggers the voice assistant, checking up on the driver’s state and takes further safety measures

Stood 4th nationwide in the Honda Hackathon

Tags: IMU, NLP, Road Safety

A software prototype web app for demand forecasting, inventory management and food tracking using machine learning and blockchain.

Runners Up in FallFest'20

Tags: Demand Forecast, Blockchain, Flask

Developing a Husky Robot with an Arm Manipulator to assist humans in day to day and industrial tasks.

Tags: ROS, Gazebo, Moveit

A web app which takes images from various CCTV in building cameras runs a Neural Net and incase of a fire sends an alert and suggests best possible escape route to users.

Tags: Image classification, Flask

An attendance monitoring system using face recognition and liveliness detector.

Tags: Face Recognition, Flask, Tensorflow

A web app where people can deposit the lost items and the the person who lost it can claim it by passing a NLP test.

Tags: Similarity of passages, NLTK

An IoT device that can help irrigate field according to the water needs of the crop. The same device can be used to sprinkle insecticide and pesticide in regular intervals. The device uses MQTT protocol which allows us to control it from anywhwere in the world.

Winner among first years in Concetto'19 Hackathon

Tags:IoT, ESP8266, Irrigation

My Schrödinger projects. I will blame it on the online classes and Netflix.

Tags: Upcoming, PROcrastinating