My work

In this section, I would like to briefly share the projects that were developed during my studies both in the Bachelor's and Master's programs as well as personal projects outside of the university. I would also like to mention the programming languages that we worked with throughout the learning process. Below is an overview of my educational journey.

Website development

During my university studies, we learned the fundamentals of web development, including HTML, CSS, JavaScript, and MySQL. As part of the coursework, we built simple user interfaces and connected them to databases, creating small-scale student projects. Although these projects are no longer available, the experience gave me a solid understanding of how core web technologies work together. Outside of the academic program, I completed an intensive 6-month web development course at CompNet.kz. The course covered the full web development cycle, including both frontend and backend technologies. We explored new topics every day and immediately applied them in practice: developing both client-side and server-side components of web applications, debugging code, and learning to independently solve real-world problems. Particular attention was given to working with interfaces, handling requests, managing databases, and understanding the principles of building stable and scalable web systems.

Yii2

I participated in creating a website for a beauty salon using the Yii2 framework. The website had two main parts: a client side and an admin panel. On the client side, users could view products and book services. In the admin panel, it was possible to add and edit products, manage availability, and see client requests. All data was stored in a MySQL database. The project was built using PHP, HTML, CSS, and MySQL.

Development of a VR Application for PC Assembly

This project is an educational Game in which the user can virtual assembly a personal computer using VR glasses and a controller. This is especially useful for beginners who want to build a PC with their own hands. After launching the application, the user enters a virtual room with a laptop. With the help of a VR controller, he can move around the room, interact with objects, select components and assembly Pc step by step. there is a laptop in the room and there you can select the necessary parts, processor, graphics card, RAM and other components. After the selection, all the items appear on the desktop. The user can take them with virtual hands (via the controller) and assemble the PC by following the instructions on the screen. After the build is completed, the virtual computer "turns on" and the game ends. Technologies used: Blender – for modeling 3D objects (graphics card, processor, case, etc.). Unity - for building the interactive environment. C# - for implementing logic and interaction. VR Headset (connected to a smartphone) - for user control and immersion.

Image coloring based on deep learning.

The Master's thesis project is dedicated to the task of automatic coloring of black and white images using deep learning methods. The main idea of the project is to develop and train a neural network model capable of reconstructing color information based on the structure and content of a black-and-white photograph. The input for the model is a grayscale image, a black-and—white photo, and the output is a colorized version of it. To train the model, the COCO-Stuff dataset was selected, which consists of 116,000 color photographs of various resolutions. Before sending all the images to the model, they were reduced to a single size of 256×256 pixels, and then divided into training and test samples. The model was trained in the Jupyter Notebook environment using Python and libraries such as PyTorch, NumPy, OpenCV, Matplotlib and other machine learning and computer vision tools. In the process, the architecture of the U-Net type model was implemented, which consists of an encoder and a decoder: the encoder consists of 7 blocks, and the decoder consists of 3 blocks. Additional improvements were also added to the model, including attention mechanisms (CBAM), and quality assessment metrics were used for comparison.: PSNR, SSIM, LPIPS. After training, the model was tested on several datasets to evaluate its versatility and color accuracy. The model showed good results by successfully coloring images that were not used in training.

Little bit about me

Hi! My name is Beksultan, and I am glad to welcome you to my page here you will learn more about me, my projects and professional interests. Thanks for stopping by!

Education: In 2018, I received a grant to study at the International Kazakh-Turkish University named after Khoja Akhmet Yasawi (IKTU) in the field of Computer Engineering: computer technology and software. In 2023, I successfully completed my bachelor's degree and won another grant for a master's degree at the Black Sea Technical University (Karadeniz Technical University, Turkey), where I am currently studying in the same specialty and working on a graduation project.

Professional interests. Software development and artificial intelligence and computer vision (my master's project is dedicated to automatic coloring of black and white photos using deep learning). Web technologies and VR applications - I like to create solutions that make life more convenient and interesting.Over the years of my studies, I have become accustomed to constant development and self-education, regularly learning new programming languages, frameworks, and tools. I believe that the most important thing in the engineering profession is the ability to learn quickly and put knowledge into practice. Thank you for your time and attention!

Hi, I'm B e k s u l t a n From Kyrgyzstan

My work

Website development

Yii2

Development of a VR Application for PC Assembly

Image coloring based on deep learning.

Little bit about me