Contact
online training
Advanced
Computer Vision
GAN, SSD, Neural Style Transfer
and Other Modern Technologies
Master modern computer vision technologies — from object detection (SSD) to image generation (GAN) and style transfer — in practice with Keras and TensorFlow. Discover technologies that change the world!
No complex mathematics
only practice and ready-made solutions.
We analyze real cases
diagnostics, stylization, generation.
Full path
from CNN to GAN
Expert support
from practitioners with extensive experience
About the course
This course is a practical intensive course on modern computer vision, where you will learn not just how to use neural networks, but how to implement them in real projects.
Where these skills are used
Expert support
Medical image analysis (X-ray, MRI, microscopy)

Automatic diagnostics (search for tumors, blood cell analysis)

Development of surgical robots and monitoring systems
Automotive industry and robotics
Driverless cars (detection of pedestrians, road signs, other cars)

Industrial robots (navigation, quality control on the conveyor)

Drones (mapping, monitoring of territories)
Security and video surveillance
Face and object recognition (airports, smart cities)

Behavior analysis (detection of anomalies, tracking of suspicious actions)

Biometrics (access systems by face or iris)
Entertainment and
creative industries
Generation images and videos (DALL E, Stable Diffusion, Deepfake)

Augmented and virtual reality (AR/VR filters, 3D reconstruction)

Gamedev (automatic texture generation, facial animation)
Retail and e-commerce
Product recognition (cashierless checkouts, warehouse automation)

Buyer behavior analysis (eye tracking, heatmap in stores)

Virtual fitting rooms (AR fitting of clothes and accessories)
Agriculture and ecology
Crop monitoring (analysis of plant conditions from drones)

Detection of pests and diseases

Tracking animals in the wild
Who is this course for?
This course is designed for those who want to move from theory to practice in computer vision and master in-demand technologies on real projects.
Practicing developers
Python engineers and Data Scientists who want to add Computer Vision to their professional toolkit and work with modern architectures (GAN, SSD, ResNet).
Technical specialists moving to AI
Analysts, backend developers and IT engineers who want to change their specialization and enter the field of machine learning and computer vision.
Students and beginners in Data Science
Those who have already mastered the basics of Python and ML, but want to delve into the applied use of neural networks for working with images and videos.
Course requirements
To complete the course, you need knowledge of Python (NumPy, Pandas), the basics of neural networks and CNN, and basic mathematics. Experience with TensorFlow/Keras is desirable. Technically, you need a PC with a GPU or access to Google Colab.
You will learn
Create and train modern neural network architectures (VGG, ResNet, Inception) for computer
Implement object detection using SSD and other advanced algorithms
Use generative adversarial networks (GAN) to create and process images
Perform style transfer between images (Neural Style Transfer)
Use transfer learning to quickly solve practical problems
Optimize models for work in real conditions
Our experts are practicing specialists in the
field of computer vision and deep learning
Andrey Volkov
Lead Computer Vision Engineer
Meditekh startup
DeepDiagnostics (development
of AI for radiology)
Olga Smirnova
Senior AI Researcher
NeuralArts (generative AI for
creative industries)
Dmitry Petrov
CV Team Lead
AutoVision Systems (developer
of software for driverless cars)
Ekaterina Kozlova
Head of AI Solutions
SmartRetail Tech (computer
vision for retail)
Course Program
Module 1 Welcome
  • Introduction
  • Plan and perspective
  • How to succeed in this course
Module 2 Google Colab and Getting Setup
  • Where to get the code, notebooks, and data
  • Intro to Google Colab, how to use a GPU or TPU for free
  • Uploading your own data to Google Colab
  • Where can I learn about Numpy, Scipy, Matplotlib, Pandas, and Scikit-Learn?
  • Temporary 403 Errors
Module 3 Machine Learning Basics Review
  • What is Machine Learning?
  • Code Preparation (Classification Theory)
  • Beginner's Code Preamble
  • Classification Notebook
  • Code Preparation (Regression Theory)
  • Regression Notebook
  • The Neuron
  • How does a model "learn"?
  • Making Predictions
  • Saving and Loading a Model
  • Suggestion Box
Module 4 Artificial Neural Networks (ANN) Review
  • Introduction
  • Forward Propagation
  • The Geometrical Picture
  • Activation Functions
  • Multiclass Classification
  • How to Represent Images
  • Color Mixing Clarification
  • Code Preparation (ANN)
  • ANN for Image Classification
  • ANN for Regression
Module 5 Convolutional Neural Networks (CNN) Review
  • What is Convolution? (part 1)
  • Part 2
  • Part 3
  • Convolution on Color Images
  • CNN Architecture
  • CNN Code Preparation
  • CNN for Fashion MNIST
  • CNN for CIFAR-10
  • Data Augmentation
  • Batch Normalization
  • Improving CIFAR-10 Results
Module 6 VGG and transfer learning
  • Introduction
  • What's so special about VGG?
  • Transfer Learning
  • Relationship to Greedy Layer-Wise Pretraining
  • 2 Approaches to Transfer Learning
  • Transfer Learning Code (Part 1)
  • Part 2
  • VGG Section Summary
Module 7 ResNet (and Inception)
  • Introduction
  • ResNet Architecture
  • Transfer Learning with ResNet in Code
  • Blood Cell Images Dataset
  • How to Build ResNet in Code
  • 1x1 Convolutions
  • Optional: Inception
  • Different sized images using the same network
  • ResNet Section Summary
Module 8 Object Detection (SSD/RetinaNET)
  • Introduction
  • Object Localization
  • What is Object Detection?
  • How would you find an object in an image?
  • The Problem of Scale
  • The Problem of Shape
  • SSD Tensorflow Object Detection API (Part 1)
  • Part 2
  • SSD for Video Object Detection
  • Optional: Intersection over Union & Non-max Suppression
  • SSD Section Summary
Module 9 Neural Style Transfer
  • Introduction
  • Style Transfer Theory
  • Optimizing the Loss
  • Code Part 1
  • Part 2
  • Part 3
  • Style Transfer Section Summary
Module 10 Class activation maps
  • Theory
  • Code
Module 11 GANs (generative adversarial networks)
  • Theory
  • Code
Module 12 Object localization project
  • Introduction
  • Localization Code Outline (Part 1)
  • Part 2/1
  • Part 2/2
  • Part 2/3
  • Part 3/1
  • Part 3/2
  • Part 4/1
  • Part 4/2
  • Part 5/1
  • Part 5/2
  • 6/1
  • 6/2
  • 7/1
  • 7/2
Module 13 Keras and Tensorflow 2 basics Review
  • Tensorflow Basics
  • Tensorflow Neural Network in Code
  • Keras Discussion
  • Keras Neural Network in Code
  • Keras Functional API
  • How to easily convert Keras into Tensorflow 2.0 code
Module 14 Course Conclusion
  • What to Learn Next
Module 15 Appendix/FAQ
  • What is the Appendix?
Module 16 Customizing your environment
  • Pre-Installation Check
  • Anaconda Environment Setup
  • How to install Numpy, Scipy, Matplotlib, Pandas, IPython, Theano, and TensorFlow
Module 17 Additional help with Python programming for beginners (as requested by students)
  • How to Code by Yourself (Part 1)
  • Part 2
  • Proof that using Jupyter Notebook is the same as not using it
  • Python 2 vs Python 3
  • How to use Github & Extra Coding Tips (Optional)
Module 18 Effective learning strategies for machine learning
  • How to Succeed in this Course (Long Version)
  • Is this for Beginners or Experts? Academic or Practical? Fast or slow-paced?
  • Machine Learning and AI Prerequisite Roadmap (Part 1)
  • Part 2
Module 19 Bonus
  • BONUS
Practice: Real projects that
you will realize on the course
This course is designed for those who want to move from theory to practice in computer vision and master in-demand technologies on real projects.
Medical blood cell analyzer
Smart parking lot with car detection
Fake Face Generator
AR “Artist” filter for selfies
Social Distance Detector
A quality control system for
manufacturing
Certificate
Upon successful completion of the course, you will receive an official certificate recognizing your skills in computer vision and deep learning, which will open access to promising career opportunities in the rapidly growing field of artificial intelligence.
Select the correct tariff
Introductory
$85
  • 3 modules
  • Video lessons
  • Practical assignments
  • Assignment Check
  • Chat for students and tutors
  • Access to the course - 1 month
  • No certificate
Start now
Basic
$120
  • 14 modules
  • Video lessons
  • Practice assignments
  • Assignment check and recommendations
  • Chat for students and tutors
  • Access to the course - 8 months
  • Certificate
Start now
Standard
$250
  • 19 modules
  • Video lessons
  • Practice assignments
  • Assignment Checks and Guidance
  • Chat for students and tutors
  • Access to the course - 12 months
  • Certificate
Start now
VIP
$400
  • Personalized support
  • Troubleshooting and recommendations
  • 19 modules
  • Video lessons
  • Practice assignments
  • Chat for students and tutors
  • Access to the course - 12 months
  • Certificate
Start now
Corporate
$670
  • Groups of 5 to 10 people
  • 19 modules
  • Video lessons
  • Practical assignments
  • Assignment check and recommendations
  • Chat for students and tutors
  • Access to the course - 12 months
  • Certificate
Start now
Testimonials from course alumni
Alexey
Data Scientist at Sberbank
Thanks to the course I was able to switch from classical ML to Computer Vision. After 2 months I implemented defect detection in production - the company saved 500k rubles/month. It was especially valuable that they analyzed industrial cases, not toy datasets.
Mikhail
Embedded developer
I was looking for a course where they give not only the theory of GANs, but also their optimization for hardware. I learned how to compress models for drones - now they analyze fields 3 times faster. The portfolio of projects immediately attracted HR.
Elena
Product manager at Wildberries
As a non-programmer, it was important for me to understand the possibilities of CV for e-commerce. After the course I initiated a project on automatic sorting of goods - we reduced errors in the warehouse by 40%. Now I lead the AI-solutions department.
Dmitry
Student
I took all possible CV MOOCs, but only here I learned how to really debug models. The final project on mask detection was included in my research, which was published on NeurIPS.
Artem
Python developer
For a long time I doubted whether my skills would be enough for CV. The course gave me a clear roadmap: from basic CNNs to complex architectures. Already after a month I made an MVP for license plate recognition - now it's part of the parking system in our shopping center. Keras + ready-made templates = perfect for pumping skills without too much theory.
Victor
C++ developer
We needed to adapt YOLO to our industrial cameras. Thanks to the block on model optimization I reduced inference time from 200 ms to 25 ms! The authors of the course really know the pain points of programmers - the code from the lectures can be immediately taken into production.
Oleg
Team Lead in IT-outsourcing
A team of 5 juns took the course. Result: in 2 weeks they built a prototype for a client - analysis of defects on the assembly line. Now we require a certificate of completion from all new candidates to the CV department.
Questions
What level of training is needed to start?
The course is designed for students with basic knowledge of Python (functions, classes, working with libraries), understanding of machine learning basics (what neural networks are and how they are trained) and minimal experience with NumPy/Pandas - if some skills are lacking, we recommend taking courses in these areas, and then continue learning here.
What hardware is required for training?
A laptop with 8+ GB of RAM will do for a comfortable course, but most of the practical tasks can be done in Google Colab with free access to GPU.
How much time should be devoted to the course?
The standard training regimen is 6-8 hours per week. All materials are available 24/7. You can learn at your own pace.
What makes this course better than others?
Our course is fully practice-oriented - there is no theoretical “water”, only actual knowledge and working code, and the training is led by industry experts, not theoretical professors, plus you get operational support and learn 2025 architectures, not outdated methods.
Can I take the course without a math background?
You can learn basic models without advanced math thanks to Keras' high-level capabilities, but advanced topics (GAN, neural network optimization) will require a basic understanding of linear algebra and gradient descent - we explain key concepts with practical examples without complex formulas.
What if I change my mind?
Don't worry - you have the opportunity to make a final decision within the first three lessons. If during this period you realize that the course is not right for you, we will refund the full price without any questions asked. If you change your mind later, we will refund the amount less the lessons completed.