Volkan is a goal-oriented Computer Vision Engineer with 7+ years of experience. Proficient in image processing and deep learning in real-time embedded systems. Developed, deployed and maintained deep learning models in different types of products.
MSc in Cognitive Science, 2023
Middle East Technical University
BSc in Electrical - Electronics Engineer, 2017
Middle East Technical University
Proficient in Python programming, with experience in web development, data analysis, and machine learning applications.
Experienced in C++ development, with a focus on high-performance computing, system programming, and software engineering principles.
Skilled in using Matlab for numerical computing, algorithm development, data visualization, and mathematical modeling.
Deep understanding of deep learning frameworks like PyTorch, Tensorflow, and Keras for building and training neural networks.
Experienced in image processing and numerical operations using OpenCV and Numpy libraries in Python.
Knowledgeable in optimizing deep learning models for production using TensorRT and TVM for efficient inference on various hardware.
Proficient in using GStreamer for building media-handling components and pipelines for streaming applications.
Experienced in optical system design, lens selection, and camera calibration techniques for computer vision applications.
Skilled in using Blender for 3D modeling, animation, rendering, and visual effects in multimedia projects.
Skilled in deploying AI and machine learning projects on Nvidia Jetson platforms for edge computing applications.
Experienced in implementing and optimizing AI and machine learning solutions on Rockchip platforms for embedded and IoT applications.
Proficient in FPGA design and development using Xilinx tools, including Vivado SDK, RTL, and High-Level Synthesis (HLS).
Experienced in utilizing V7 Darwin for AI-powered image annotation, data management, and model training.
Extensive experience with Linux operating systems, including system administration, shell scripting, and kernel customization.
Experienced in version control systems, particularly Git and SVN, for effective team collaboration and code management.
Skilled in project management and documentation using tools like Jira and Confluence, along with proficiency in Microsoft Office and similar programs.
The study aimed to classify task difficulties using wavelet transform images of EEG signals and deep learning models. The EfficientNet-B0 model achieved the highest accuracy, but its performance varied significantly across individuals and task difficulties, indicating limited generalizability. The study suggests a need for further research to improve model generalizability, optimize performance, and validate the models on larger, more diverse datasets.