Image generation model that you can understand by implementing it - From CNN basics to Diffusion

Let's implement a deep learning framework directly with Excel! Let's solve supervised learning/unsupervised learning/reinforcement learning with the implemented framework!

Let's learn the principles of deep learning by visualizing them with Excel.

[Lecture Summary: From Pixels to Practice, Building a Business Card Recognition System] * **Pre-processing:** We learned how to remove noise and extract the 'skeleton' of a business card using Gaussian Blur and Canny Edge detection. * **Hough Transform:** We went through an "audition" process to select the real 'borders' from tens of thousands of points using the $(\rho, \theta)$ voting system. * **Geometric Reasoning:** By solving complex linear equations through substitution and matrices, we precisely targeted the four corners $(x, y)$ of the business card. * **Data Sorting & Filtering:** We mastered data refinement techniques, such as removing overlapping lines and labeling the found points in the order of 'top-left, top-right, bottom-right, bottom-left.' * **The Magic of Transformation (Perspective):** We concluded with practical techniques for straightening tilted images or, conversely, compositing an image into a frame of your choice. 📢 "Does looking at formulas give you a headache? It was the same for me." When I first took on an image processing project in the field, what bothered me most were the screens full of $\sin$ and $\cos$ formulas. Textbooks were full of nothing but equations, and nowhere provided answers to practical problems like, **"Why can't this find the business card in my code?"** Through this lecture, I want to help you solve three chronic problems you will encounter: 1. **"The problem where math feels like an alien language"** → Solved through [The Power of Analogy]. Instead of simply memorizing linear equations, I help you understand them through **'lasers fired from the origin'** and **'shadow play.'** Once you understand the principles, the code follows naturally. I translate formulas into the 'easiest language' that I discovered after countless trials and errors. 2. **"The problem where you know the theory but can't apply it"** → Solved through [Data Auditions]. Why are there 10 borders on my business card? Why is the order of the corners scrambled? I have revealed all my debugging know-how on how to filter and sort the 'messy data' you inevitably encounter in practice. The audition logic for separating the wheat from the chaff will take your project to the next level. 3. **"The problem where the result is boring"** → Finished with [Interactive Composition]. This isn't a lecture that ends with just looking at black-and-white edges. The moment you tuck the business card you created into a 'Hall of Fame frame' that you clicked yourself with a mouse, you will experience **'the magic of manipulating space'** beyond simple coding. "Computer vision is not about math; it's the process of gaining a new set of eyes to see the world." I want to pass on to you that thrilling 'A-ha!' moment I realized in the field. You can close the complex theory books now. Let's paint images freely with the brush of code together!

This lecture examines the entire process of generating a photo mosaic from an algorithmic perspective. The input image is reduced to a set of representative colors matching the number of tiles through K-means clustering, then the Hungarian algorithm is applied to optimally match each color with a tile in a 1:1 correspondence. Through this, you will understand how image processing and combinatorial optimization algorithms combine to create a single visual output.

Python + xlwings Work Automation Course Summary 1. Basic Section: Excel Control Fundamentals Cell Handling / Format Application / Image & Chart Insertion 2. Application Section: Automation Scenarios Automatic Email Sending / Report Summary Automation (LLM) / Image Classification Tasks (Teachable Machine)

Why this course is special: Key Advantages • Intuitive Visualization: Directly observe algorithm operations in real-time with Pygame 2D simulations • Practical Implementation Experience: Go beyond theory and internalize autonomous driving algorithms by coding directly • Master Core Algorithms: Focused learning of essential algorithms such as Dijkstra, Pure Pursuit, ICP, etc. • Step-by-step Advanced Learning: Systematic difficulty progression from basics to SLAM • Lidar-based SLAM: Practical map building and localization in unknown environments

Deep learning is a technology that learns data through neural networks composed of combinations of complex functions. In this lecture, we will mathematically understand the core concepts of deep learning and analyze them from the perspective of matrix operations. In particular, utilizing Python's NumPy library, we will visually examine how parameters are updated by directly implementing the forward and backward propagation processes of deep learning. Even the seemingly complex neural network structure becomes clear when analyzed with matrix operations. This lecture focuses more on understanding concepts than coding and is suitable for students who wish to intuitively grasp the principles of deep learning mathematically.

Implementing deep learning basics with Excel to deeply understand their meaning! Learn how image data is processed!

Implement Deep Learning Basics in Excel for Deeper Understanding!

Deeply understand the meaning of deep learning by implementing its basics in Excel!

Pre-Trained Model (VGG16) / CAM / Grad-CAM / YOLO / STYLE TRANSFER / PIX2PIX / CYCLEGAN Implementing Stable Diffusion Directly

Pre-Trained Model (VGG16) / CAM / Grad-CAM / YOLO / STYLE TRANSFER / PIX2PIX / CYCLEGAN Implementing Stable Diffusion Directly

This course is a hands-on masterclass that completely dissects the core technological evolution of generative AI, from LDM (Latent Diffusion Model) to DiT (Diffusion Transformer). We directly analyze the latent space-based learning principles of LDM, the structure of Stable Diffusion, and the implementation methods of the latest Diffusion Transformer through papers and code. Students will systematically learn the latest trends and structural evolution of generative models by directly implementing LDM, CFG (Classifier-Free Guidance), and DiT models using PyTorch.

We implement the latest Transformer-based PixArt and lightweight adaptation SANA step by step from theory to code. Building on DDPM·DDIM·LDM·DiT covered in Parts I·II, we complete hands-on practice including text encoder integration, samplers (DDIM/ODE), v-prediction/CFG tuning, and small-scale data style fine-tuning.

This course conducts hands-on practice related to Diffusion models among generative artificial intelligence models. By reading and implementing the prompt-to-prompt paper, which is a representative Diffusion model application paper, we expect to cultivate the ability to understand the latest artificial intelligence papers.

Are you feeling overwhelmed by practical application and interview preparation? Based on my industry experience, I will help you firmly grasp everything from CNN to DETR, with a strong focus on code.

This lecture examines the entire process of generating a photo mosaic from an algorithmic perspective. The input image is reduced to a set of representative colors matching the number of tiles through K-means clustering, then the Hungarian algorithm is applied to optimally match each color with a tile in a 1:1 correspondence. Through this, you will understand how image processing and combinatorial optimization algorithms combine to create a single visual output.

Learn how to build a real-time fire detection system using Next.js, YOLOv11, and FastAPI.

Solve the point where 80% get stuck at makeathons. Complete Raspberry Pi · Computer Vision · LLM · 3D Design in 4 weeks! Achieve top rankings at makeathons with a demonstrable smart mirror PoC!

If you want to properly learn OCR technology that's truly used in practice, this one course is all you need! Aiming for over 98% accuracy even with unstructured documents and complex layouts, based on the latest SOTA models and real-world know-how, we build enterprise-level OCR projects together.

This is a comprehensive machine learning project course that learns various useful machine learning regression and classification projects along with theory using Python and TensorFlow 2 based on Boston Marathon big data.