Skip to content
View yaghmo's full-sized avatar
🤨
🤨

Block or report yaghmo

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
yaghmo/README.md

Yaghmoracen Belmir

AI Engineer | Deep Learning | Computer Vision | NLP | Generative Modeling

I work on end-to-end machine learning systems with a focus on computer vision, large language models, generative approaches (GANs), and scientific imaging.
My experience includes medical imaging research, applied deep learning, and building production-oriented AI applications.


Selected Work

MRI-to-CT Synthesis using GANs (LaTIM, 2025)

Generative model for synthesizing CT from MRI to support radiotherapy dose planning.

  • Developed preprocessing and paired dataset preparation pipeline
  • Trained and evaluated GAN architectures (PSNR, SSIM, structural metrics)
  • Achieved a 2.5% improvement in anatomical preservation compared to baseline

Repository / Report: coming soon
Technologies: PyTorch, GANs, Medical Imaging


Multi-Model Chatbot

A modular conversational interface supporting inference from multiple LLM backends.

  • Implemented token streaming, temperature/top-p tuning, and model switching
  • Designed multilingual handling for real-time translation and interaction
  • Built with modular architecture for future model integration

Repository: https://github.com/yaghmo/chatbot
Technologies: Python, Streamlit, Hugging Face Transformers


Medical Image Segmentation Benchmark

Comparative study of segmentation architectures for lung region extraction.

  • U-Net, DeepLabV3, and FCNN under identical training and evaluation settings
  • Evaluation using Dice, IoU, and boundary metrics

Repository: https://github.com/yaghmo/U_Net-DeepLabV3-FCNN-lung-segmentation
Technologies: PyTorch, Computer Vision


Object Detection and Tracking with Spatial Relation Modeling

Real-time pipeline combining YOLOv8 detection with SORT tracking.

  • Extraction of spatial interaction descriptors (distance, direction, relative motion)
  • Framework supports downstream tasks such as scene understanding or activity prediction

Repository: https://github.com/yaghmo/yolov8-sort-detection-tracking-segmentation-force-histogram-banner-cnn
Technologies: YOLOv8, SORT, NumPy, OpenCV


Multilingual Sentiment Analysis System

End-to-end sentiment analysis workflow integrating language identification, translation, classification, and LLM-based summarization.

Repository: https://github.com/yaghmo/sentiment_analysis
Technologies: Python, Transformers, NLP


Research Interests

  • Computer Vision and Image Understanding
  • Medical Imaging and Scientific Computing
  • Generative Modeling (GANs, diffusion models)
  • Large Language Models and conversational AI
  • Multimodal AI systems

Technical Competencies

Machine Learning:
Deep Learning, Computer Vision, NLP, Generative Models, LLMs, Representation Learning

Frameworks:
PyTorch, TensorFlow, Hugging Face, MONAI, Scikit-Learn, OpenCV

Systems & Tools:
Docker, Git, Linux, Streamlit

Programming:
Python, C, C++


Contact

LinkedIn: https://www.linkedin.com/in/belmir-yaghmoracen/
Email: yaghmo.belmir@gmail.com

Pinned Loading

  1. chatbot chatbot Public

    Flexible conversational AI system capable of switching between different LLMs and supporting multilingual users

    Python

  2. sentiment_analysis sentiment_analysis Public

    Sentiment pipeline usable by global users, regardless of language or dialect

    Python

  3. yolo-sort-spatial-relations yolo-sort-spatial-relations Public archive

    Real-time object detection and tracking pipeline combining YOLOv8 and SORT, with extraction of spatial interaction features for scene understanding.

    Jupyter Notebook 1

  4. U_Net-DeepLabV3-FCNN-lung-segmentation U_Net-DeepLabV3-FCNN-lung-segmentation Public archive

    This project compares U-Net, FCN, and DeepLabv3 models for instance-level lung image segmentation

    Jupyter Notebook

  5. Image-to-text-to-speech Image-to-text-to-speech Public archive

    Extract text from images, correct it using NLP, and synthesize it into audio.

    Jupyter Notebook 2