QAT(quantize aware training) for classification with MQBench

This is a project documentation about melanoma detection methods using convolutional neural networks.

🔪 Elimination based Lightweight Neural Net with Pretrained Weights

American Sign Language Alphabet Detection in Real Time using OpenCV-Mediapipe with EfficientNetB0 in PyTorch

An implementation of the Arabic sign language classification using Keras on the zArASL_Database_54K dataset

Development of a depth estimation model based on a UNET architecture - connection of Bi-directional Feature Pyramid Network (BIFPN) and EfficientNet.

TFlite+CNN

The purpose of Food Vision project is to classify 101 variety of food items using Machine Learning.

SkinNet Analyzer: A Deep Learning-Based Skin Disease Detection System - College Final Year (4th year) Project

A multi classification using scikit-learn and TensorFlow models on MRI scans of patient's brains.

Image Captioning using EfficientNet and GRU

Dust detection on solar photovoltaics panel using pre-trained CNN models

CoalClassifier: A deep learning model for classifying coal types using EfficientNetB0-based transfer learning and fine-tuning techniques. This project is designed to accurately distinguish between Anthracite, Bituminous, Lignite, and Peat classes and is developed using TensorFlow/Keras

HAM10000 Skin Lesion Classification

Deepfake detection leveraging the OpenForensics dataset, a comprehensive resource for face forgery detection and segmentation research. The project explores various deep learning models and evaluates their performance in distinguishing between real and fake images of human faces.

49.5 mAP50 Detector enet4y2-coco.cfg = EfficientnetB0 + 4YOLO Layers + BiDirectionalFeatureMap with COCO Dataset and 81.0 mAP50 with VOC2007 test Dataset.

A Deep Learning application for Malaria Detection

Mask Monitoring System

Efficient way to detect if a face has a mask on. Used EfficientNet-B0 model architecture.

A Streamlit aaplication to distinguish between AI-generated and real images. Utilizing the TensorFlow framework, the core model is based on EfficientNetB0, which has been fine-tuned on a custom dataset containing both fake and real images.