This project focuses on implementing CNN model based on the EEGNet architecture with Pytorch library for classifying motor imagery tasks using EEG data.

Reference implementation for Blueprint Separable Convolutions (CVPR 2020)

Keras w/ Tensorflow backend implementation for 3D channel-wise convolutions

Sound event detection with depthwise separable and dilated convolutions.

This code implements the EEG Net deep learning model using PyTorch. The EEG Net model is based on the research paper titled "EEGNet: A Compact Convolutional Neural Network for EEG-based Brain-Computer Interfaces".

Code for "Complex-Valued Depthwise Separable Convolutional Neural Network for Automatic Modulation Classification"

Online learning platform with automatic engagement recognition

Efficient Deep Learning for Real-time Classification of Astronomical Transients and Multivariate Time-series

Cheng-Hao Tu, Jia-Hong Lee, Yi-Ming Chan and Chu-Song Chen, "Pruning Depthwise Separable Convolutions for MobileNet Compression," International Joint Conference on Neural Networks, IJCNN 2020, July 2020.

Xception V1 model in Tensorflow with pretrained weights on ImageNet

PyTorch implementation of Depthwise Separable Convolution

MobileNet V2 transfer learning with TensorFlow 2.

"Advanced Machine Learning" project @ Politecnico di Torino, a.y. 2021/2022.

A novel architecture for enhancing image classification. Reference paper: https://arxiv.org/abs/2104.12294

Implementation of state-of-the-art models to do segmentation over our own dataset.

A TensorFlow2.0 implementation of Xception Deep Learning with Depthwise Separable Convolutions

I Implemented some of the custom complex Convolutional Neural Network architecture using tensorow.keras Functional API.

Smart Automation Controller for Precision Agriculture

Neural Network for Low Complexity Acoustic Scene Classification

Project crafted by Antonio Ferrigno, Giulia Di Fede and Vittorio Di Giorgio for the Advanced Machine Learning course at Politecnico di Torino (2023/2024)

Performance Evaluation between Normal and Depthwise Seperable Convolutions for Medical Image Classification.

Project for "Advanced Machine Learning" course at PoliTO. The purpose is to implement a BiSeNet able to perform real-time semantic segmentation task

Compare regular CNN with depthwise separable CNN for lightweight network

Exploiting Two-Level Pipeline and Triple Sliding-Window Input Buffer for Efficient DSC Accelerator Design

Lightweight Attention U-Net for Breast Cancer Semantic Segmentation

Implementation and research paper of MobileNet

The preparation for the Lung X-Ray Mask Segmentation project included the use of augmentation methods like flipping to improve the dataset, along with measures to ensure data uniformity and quality. The model architecture was explored with two types of ResNets: the traditional CNN layers and Depthwise Separable.

Exploiting Two-Level Pipeline and Triple Sliding-Window Input Buffer for Efficient DSC Accelerator Design

This repository contains research on real-time domain adaptation in semantic segmentation, aiming at bridging the gap between synthetic and real-world imagery for urban scenes and autonomous driving, utilizing STDC models and advanced domain adaptation methods.