Skip to content
View metin-yat's full-sized avatar
0 followers · 1 following

Block or report metin-yat

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
metin-yat/README.md

💫 About Me:

I am a fourth-year Statistics student at Gazi University, specializing in Computer Vision and the development of production-ready AI applications. My statistical background provides a robust analytical foundation for deep learning, high-dimensional data analysis, and model optimization.

Having transitioned from full-stack web development (Django/React) to AI engineering, I now focus on bridging the gap between academic research and scalable automation. My current R&D efforts are centered on 3D Reconstruction, specifically working with Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (GSplat) to build intelligent automation pipelines.

Moving forward, I intend to deepen my focus on 3D Vision, specifically targeting the intersection of 3D Reconstruction and Generative Modeling. My goal is to contribute to the field through research-driven projects that explore the generation of complex 3D environments and the advancement of spatial intelligence.

Tech Skills

  • Computer Vision: 3.5+ years
  • Machine Learning: 4+ years (Statistical Modeling, XGBoost, Catboost)
  • 3D Vision & Research: NeRF, GSplat, Gaussian Splatting (custom 3DGS implementations)
  • Language Models: LLM Orchestration and Vision-Language Models (VLM)

  • Languages: Python (5+ years), C/C++, R, MATLAB
  • DevOps and Tools: Docker, Linux (Ubuntu), CUDA, AWS (Basics)
  • Robotics: ROS and Gazebo integration
  • Backend: Django, Flask, SQL

  • Frontend: React.js
  • Design: Blender (Beginner)

Pinned Loading

  1. vision-risk-detection vision-risk-detection Public

    A decoupled asynchronous pipeline utilizing a producer-consumer architecture to bridge real-time object detection with VLM-based reasoning.

    Python

  2. 2d-gaussian-splatting 2d-gaussian-splatting Public

    A PyTorch-based differentiable 2D Gaussian rasterizer for image reconstruction from scratch.

    Python

  3. yolov8-tensorrt-optimization yolov8-tensorrt-optimization Public

    YOLOv8n model training, FP16/INT8 quantization, and TensorRT pipeline with performance benchmarking.

    Jupyter Notebook

  4. mini-splat mini-splat Public

    minimal 3d gaussian splatting framework. implements differentiable rendering, 3d-to-2d projections, and point cloud densification from scratch.

    Python

  5. Geospatial-Lidar-Project Geospatial-Lidar-Project Public

    Jupyter Notebook

  6. model-library model-library Public

    Python