This project aims to classify different types of fruits using deep learning. The objective is to build a model that can accurately identify the type of fruit based on images.

Built a deep learning model for flower image classification using the pre-trained MobileNetV2. By fine-tuning this lightweight CNN, the system accurately identifies flower species with efficient performance, making it suitable for real-time and resource-constrained environments.

Polyp segmentation on the Kvasir-SEG dataset using deep learning models like UNet, UNet++, PSPNet, and DeepLabV3Plus with ResNet50 and MobileNetV2 encoders.

An AI-powered web application for early detection of skin cancer, burns, and infections using deep learning (MobileNetV2). Trained on diverse medical datasets, the system achieved 90%+ accuracy and provides real-time predictions through a user-friendly interface.

This repository contains a Python application for emotion recognition using facial expressions. The application utilizes computer vision techniques and deep learning models to analyze facial expressions , aiming to detect and classify emotions based on input images or video streams.

🧠 Detect brain tumors from MRI scans using MobileNetV2 deep learning! 86.6% accuracy with transfer learning. Complete EDA, visualization, and production-ready code.

This is an official pytorch implementation of WideHRNet: an Efficient Model for Human Pose Estimation Using Wide Channels in Lightweight High-Resolution Network.