Hi, I'm Kevin Depedri 👋

I am a passionate engineer, working in the AI/ML field.

I am a graduate from the University of Trento 🎓

There I earned a Bachelor's degree in Information and Communications engineering 📡

and a Master's degree in Artificial Intelligence Systems 🤖

with a specialization in Computer Vision and Augmented Reality 👓

Here you can find some of the projects on which I have worked in the last years 💻

Feel free to contact me here on github for any project related question or collaboration.

Otherwise, you can reach me on Linkedin here 📫

Kevin Depedri's Projects

ai-for-finance

Stock market prediction on 5 italian companies using VAR model, OLS regressions and LSTM recurrent neural networks over data retrieved from Refinitiv Eikon

android-studio-examples

Small Android apps built in Kotlin following the Android Basics with Compose guide provided by Google

automated-planning

Resolution of an increasingly complex planning problem leveraging different PDDL and HDDL planners to achieve optimal plans

collect-and-display-a-dataset-in-augmented-reality-scenario

Android app based on Unity and ArCore to acquire 2D holograms, host their pose in the cloud and retreive them from the cloud when needed

face-spoofing-detection-using-colour-texture-analysis

Implementation of the approach presented in the paper 'Face Spoofing Detection Using Colour Texture Analysis' to tackle spoofing attempts

kevindepedri

Config file for my github profile

mpi-parallel-closest-pair-of-points

Parallel MPI implementation of Bruteforce and Divide et Impera approaches to solve the Closest Pair of Points problem leveraging the HPC infrastructure provided by the University of Trento

novel-class-discovery

Testing AutoNovel and UNO novel-class-discovery techniques to assess their prerformances under different settings. Plot of tSNE to evaluate clustering capabilities for known and new classes

sentiment-analysis

Subjectivity removal and Polarity classification of movie reviews employing a shallow model (Multinomial Naive Bayes) and a deep model (Bidirectional LSTM with self-attention)

spatial-temporal-calibration-for-outdoor-location-based-augmented-reality

Official repository of the paper "Spatial-Temporal Calibration for Outdoor Location-Based Augmented Reality" authored by Lorenzo Orlandi, Kevin Depedri and Nicola Conci