Name: Alberto Baldrati
Type: User
Company: University of Florence - MICC, University of Pisa
Bio: Ph.D. student at Media Integration and Communication Center (@miccunifi ) - University of Pisa/University of Florence
Twitter: A_Baldrati
Location: Florence, Italy
Blog: https://abaldrati.github.io
Alberto Baldrati's Projects
In AugmentBrain we investigate the performance of different data augmentation methods for the classification of Motor Imagery (MI) data using a Convolutional Neural Network tailored for EEG named EEGNet.
[ACM TOMM 2023] - Composed Image Retrieval using Contrastive Learning and Task-oriented CLIP-based Features
[CVPR 2022 - Demo Track] - Effective conditioned and composed image retrieval combining CLIP-based features
https://reddeadrecovery.gitlab.io/dawntime/ In dawntime we implement a volumetric light scattering effect based on the postprocessing technique described by Kenny Mitchell.
In facestretch we describe how we exploited the dlibβs facial landmarks in order to measure face deformation and perform an expression recognition task. We implemented multiple approaches, based metric learning, neural networks and geodesic distances
In MT-BERT we reproduce a neural language understanding model which implements a Multi-Task Deep Neural Network (MT-DNN) for learning representations across multiple NLU tasks.
In SupeRAuGAN we implement a novel data augmentation technique tailored to Generative Adversarial Networks in order to reduce discriminator overfitting and stabilize training