Vasileios Saveris's repositories
trading-technical-indicators
Trading Technical Indicators python library
common-data-model-schema
Schema extraction from Microsoft CDM files, in python.
lunar-lander-DQN
Implementation of a Deep Reinforcement Learning agent (Deep Q-Network) for landing successfully the ‘Lunar Lander’ from the OpenAI Gym.
application-server-traffic-forecasting
Traffic forecasting study case of an application server (cluster, servers x4) using Machine Learning.
deep-learning
Deep Learning projects inspired from my studies in the Deep Learning nanodegree (Udacity).
effectivepython
Effective Python: Second Edition — Source Code and Errata for the Book
graph-algorithms
Implementation of graph algorithms in Python.
opencv-detect-faces
Use of OpenCV's implementation of Haar feature-based cascade classifier, to detect human faces in images.
poisson-image-editing
'Poisson Image Editing' by Patrick Perez Michel Gangnet and Andrew Blake (reference [1]) is a research paper in which a technique for seamless blending images together is presented. 'The mathematical tool at the heart of the approach is the Poisson partial differential equation with Dirichlet boundary conditions which species the Laplacian of an unknown function over the domain of interest, along with the unknown function values over the boundary of the domain.' The main scope of this project is to implement in python the mathematical tool described in the paper. Additionally, I replicate some of the show cases presented in the paper and I compare my results with the originals. Finally, I apply the developed method in photos taken by my own and images taken from the web.
raspberry-pi-vehicle-particle-filter
Particle filter implementation on a raspberry-pi home made vehicle.
road-sign-detection
Road sign detection is a technology started being available in public already in 2008 (BMW 7 series) and it is continuously evolving. Currently its main use is as a component of the advanced driver assistance systems (ADAS) and a main feature on the design of autonomous cars. In the scientific field, there are a lot of papers from researchers in which they describe and demonstrate work flows for a more accurate and faster Traffic Sign Detection. In their focus it is not only to suggest improvements in the traditional computer vision techniques but additionally to combine Machine Learning algorithms which can boost the performance of object detection in general. In this project, my aim is to study the recent papers on the topic and to produce a system which can successfully recognize Traffic Signs and Lights from real world images. My selected approach and implementation results will be presented on this report.