Language Breakdown
Lines of code distribution across 19 owned repositories
T-Shaped Developer
T-shapedDeep in Python with broad versatility
Collaboration Network
Global Impact visualization
Repos
26
PRs
0
Growth
+18%
Top Collaborators
No collaborator data yet.
Coding Streak
Contribution activity over the past year
TRS
@TheRobotStudio
Elsa
@elsamoreno
Oğuzhan Sarıgöl
@oguzhansarigol
fabiomontello
@fabiomontello
Lucas Sandby
@Lucasmogsan
Top Repositories
MedVision Reconstructor is a Python-based solution for medical image reconstruction using advanced Radon transforms and iterative algorithms (FBP, SART). It features a PySide6 GUI for visualization of sinograms and reconstructed images, facilitating seamless testing and analysis.
Flask-based itinerary planner for Istanbul's rail lines. Includes route search, A* pathfinding, and a Leaflet-based map.
This repository contains code for the DTU course 34755 Building Dependable Robot Systems, in which we design, construct, and program a robot to compete in a obstacle course competition.
Multi-sensor multi-target tracking system for harbour surveillance. Fuses mm-wave radar, stereo camera, AIS and GNSS via an Extended Kalman Filter with Hungarian (GNN) data association and M-of-N track lifecycle. Validated on simulated scenarios and real Copenhagen harbour recordings.
Loosely-coupled GNSS/INS sensor fusion in Python: Single Point Positioning (SPP), IMU noise characterization (Allan variance, PSD), and an Extended Kalman Filter for integrated navigation.
Multimodal image geolocalization via contrastive retrieval. Embeds ground photos, satellite imagery, text and GPS into a shared space (GeoCLIP-inspired), with cross-view retrieval and University-1652 transfer experiments.
Graph Neural Networks for molecular property prediction on QM9, with semi-supervised training (Mean Teacher & N-CPS) in PyTorch Geometric
End-to-end MLOps pipeline for 5-class semantic segmentation of drone imagery using nnU-Net (PyTorch). Features DVC, Docker, BentoML serving & CI/CD.
A Python-based character recognition system using Hu, R, and Zernike image moments. Features a Tkinter GUI for easy image selection and recognition. Current development focuses on improving prediction accuracy and expanding the feature database.
This project reads a custom “drawing language” from a file or text input, uses regular expressions for lexical analysis (tokenization), then applies a context-free grammar style parser to validate syntax (supporting nested loops, color changes, and pen thickness). Finally, the parsed commands are rendered via SFML/TGUI.
Open Source Impact
Contributions to external projects
