igeor Goto Github PK

antidote-recommendation-for-health

This repository contains code for an Antidote Recommendation Tool that utilizes vector-based and topic modeling approaches to recommend antidotes for fake claims. In addition the tool leverages the bipartite graph representation and topic modeling techniques to identify relevant antidotes from a corpus of claims.

cardiotocographic-signal-classification-based-on-heart-rate-variability

compiler_starlett

Compiler development using Python

cubegame

ddpm-cl

elemental-distribution-mapping-with-deep-learning-methods

Master Thesis: Applying Deep Learning Methods for Elemental Maps Prediction. We incorporate Physics-Based Prior Knowledge to enhance the DNNs accuracy. Specifically, we introduce PIPL (Phisics Informed Prior Layer) with constant weights - the elemental spectral signatures.

graph-based-image-segmentation

This repository contains the implementation of a graph-based image segmentation algorithm. It was developed as an assignment for a computer vision post-graduate course.

igeor.github.io

image-text-removal-with-generative-adversarial-networks

Diploma Thesis: Inpainting based image text removal using Conditional GANs. It explores supervised training, and additive text localization networks, with a focus on attention gates for enhanced inpainting results.

seam-carving

Seam carving (or liquid rescaling) is an algorithm for content-aware image resizing.

texttospeecheditor

The objective of this project is to develop a simple editor that allows preparing text documents and transforming them to audio. Such an application can be much help for people who have speaking problem. Moreover, it could be a useful tool for preparing presentations that are automatically played without a physical speaker. The application consists of a graphical user interface, a back-end that enables the text editing process and the part transforms text to speech , which will be based on external text to speech libraries.