An automated plant disease detection, Progressive Web App that will help the farmers to detect the disease in their crops and will also give insights on its treatment .

Web app to recognize 3,755 Chinese words

KisanSahayak is an AI-powered, end-to-end smart agriculture solution that provides farmers with data-driven insights via web and mobile apps. Using advanced machine learning, it offers rainfall analysis, crop recommendations, and disease prediction to help optimize yields and improve crop health. 🌾

MERN Stack Web Application "EpiDetect" which uses a fine-tuned ResNet50 model for skin disease detection.

PredictiX is a comprehensive multi-disease prediction system built using the MERN stack and integrated with machine learning models. It accurately predicts lung cancer, breast cancer, diabetes, and heart disease, providing a seamless user experience for health diagnostics.

📕 Deep-learning trained Pokédex.

Emotion Recognition Project Adapted in HTML/Javascript/CSS with Tensorflowjs

A website that uses deep learning and webcam data to detect bad desk habits || Won Most Scalable in Hack the Lib 2020

GreenGuard is a multilingual AI-powered chatbot for crop disease detection, integrating AGDFNet Model. It enables farmers to upload plant images and receive real-time, text-based treatment recommendations. Using Adaptive Lesion Modules and Feature Aggregation, it enhances disease detection, helping mitigate crop losses efficiently.

SmartPosture: Personalized Study Posture Care

KisanSahayak is a smart agriculture web application aimed at providing Indian farmers with data-driven insights using advanced machine learning, rainfall analysis, crop recommendations, and disease prediction.

A Simple CNN Demo in JavaScript

PredictiX is a comprehensive multi-disease prediction system built using the MERN stack and integrated with machine learning models. It accurately predicts lung cancer, breast cancer, diabetes, and heart disease, providing a seamless user experience for health diagnostics.

VKontakte app for ambient sound recommendation using given wall post.

Presentation on Normalizer-Free (Res)Nets conducted at Wrocław University of Science and Technology on 10 March. 2021.