Join us for an immersive workshop, where you'll learn how to leverage synthetic data to accelerate your computer vision product cycles. In the age of data-centric AI, the largest potential for computer vision optimization lies in data preparation. Physical data collection and processing can be costly and limited in variability, while synthetic data opens new horizons for shorter product launch cycles and improved model robustness.
During this workshop, our experienced instructors will guide you through the process of using synthetic data for computer vision development. You'll gain a deep understanding of the benefits and limitations of synthetic data compared to physical data. We'll explore typical processes and tools for synthetic data generation, with a particular focus on NVIDIA Omniverse and how its Replicator tool can help generate visual data for model training.
The workshop will feature an interactive session where you'll dive into a real-world case study and engage in ideation exercises. This hands-on experience will allow you to apply the concepts learned and explore the potential of synthetic data in your own computer vision projects.
By the end of this workshop, you'll:
Understand the importance of data challenges in computer vision development
Explore how NVIDIA Omniverse and Replicator can facilitate visual data generation for model training
Recognize the benefits and limitations of physical data versus synthetic data
Apply the concepts learned to a case study and engage in ideation exercises
Gain knowledge of typical processes and tools for synthetic data generation
Summarize the key takeaways and understand the significance of accelerating product cycles with synthetic data
Agenda and timing
10 mins
Introduction
Get acquainted with participants and instructors;
Discuss workshop objectives and what you can expect to learn;
Outline the agenda for the session.
5 mins
Data challenges in computer vision development
5 mins
Physical data vs Synthetic data: benefits and limitations
5 mins
Synthetic data generation: typical processes and tools
5 mins
What is NVIDIA Omniverse and how Replicator can help generate visual data for model training
20 mins
Interactive session: Case study and ideation
Takeaways
Implementation process checklist
Two examples
Sample architectures
Our speakers
Julia Rubtsova
Data product manager / NLP solution architect
Ph.D. in Data Science. For the last four years with Data Monsters, I am responsible for launching, growing, and developing complex, knowledge-intensive IT products that use Data Science and Machine Learning technologies.
German Suvorov
Business Development - NVIDIA Ecosystem, Siemens
An expert in AI for industrial applications with over 20 years of experience in automotive manufacturing, advanced materials engineering, industrial automation, supply chain management, and information technologies.
Max Pavlov
Head of Project Office, Data Monsters
Building a bridge between AI technologies and business needs. Putting AI into production.
Dan Lesovodski
Partner, Data Monsters
Managed 46 AI projects for large companies, including GE, Intel, AB InBev, and Boston Scientific. Author of the course “AI for managers” on Coursera.
WHY CHOOSE US?
NVIDIA is the leading provider of GPUs for AI, and scientific research
An NVIDIA Elite service delivery partner with a portfolio of 100+ AI projects.
The global market leader in industrial automation and industrial software.
Don't miss this opportunity to enhance your computer vision skills and unlock the potential of synthetic data. Join us for the Computer Vision Transformation Workshop and take a significant step towards optimizing your product cycles.
Limited seats are available, so register now to secure your spot!
This hands-on workshop will help you procure and commission an AI system for industrial quality inspections:
Determine requirements
Ask the right questions
Estimate budgets and resources
Evaluate and secure business value
Select vendors
Identify risks
Build expectations
Structure the project
Avoid pitfalls
Commission the system to pilot and production
During this session, you will work on your own case. The workshop will be held by industrial AI experts from leading tech companies.
Agenda and timing - 60 mins
Introduction - 5 mins Use case estimations & budget and risk assessment - 15 mins Participants will use Data Monsters’ framework to structure their use cases and required resources. Data volume, throughput, and storage strategies (time to store & fast and long-term storage). Develop solution architecture and determine integration strategies. Business case evaluation. Relation of AI accuracy and business value - 10 mins We will use a spreadsheet to evaluate the business effect basing on a participant’s use cases. How to choose a system. Evaluation parameters - 15 mins Requirements, expectations, and the right questions to ask. Participants and experts will elaborate the key features that define success or failure.
Project structure. AI system commissioning - 5 mins Project structure framework: POC, Pilot, and Production phases. The ins and outs, what to expect, and how to avoid pitfalls. Common mistake at commissioning - overfitting, and train\test mismatches. Your expert team roles.
Next steps - 5 mins Q&A section - 5 mins
Takeaways
Framework
Checklists
Sample process
Sample architectures
This practical session will help you to:
• Establish and clarify your understanding of AI core concepts, • Learn about best AI use-cases in your industry, • Understand what 5 areas should be managed within an AI project, • Set up your next actions towards your goal.
During the session, we will cover 8 steps and topics:
• “What is AI?”, “How does AI work?, and “What can and can't AI do?” • The most popular AI use cases in manufacturing. • How to spot a feasible and valuable AI use-case that relates to your needs? • How to estimate AI projects and evaluate ROI? The cost structure. • Assessment, PoC, Pilot - The major phases of an AI project’s life cycle. • What can go wrong? The main risks in AI projects. • Success or failure? Terminate or continue? Setting the right acceptance criteria. • Scaling challenges. How to go beyond a PoC?
Who can benefit from this session?
If you're a manager or director striving to make the most out of AI, and you work in one of these industries: •Automotive Manufacturing •Food & Beverages •Electronics & Semiconductors •Logistics & Supply Chain
• Use cases • AI building blocks • Best practices in AI project managementand understanding of how to deal with: - Practical AI implementation - AI team roles & expectations - Investment in AI: risks, deliverables, expenses - How to measure success
• Use cases • AI building blocks • Best practices in AI project managementand understanding of how to deal with: - Practical AI implementation - AI team roles & expectations - Investment in AI: risks, deliverables, expenses - How to measure success
Speakers
Russ Sagert
Director Business Development - Industrial Manufacturing / IoT, NetApp
An expert in developing and bringing to market Digital Transformation solutions for manufacturing plant operators across the oil and gas, mining, power utilities, automotive, and high-tech fabrication industries.
DJ Bush
Global Client Executive - HPC/AI/ML, Sycomp
A global client executive working closely on integrating AI/ML solutions across different industries.
Piyush Modi
Business Development of Industrial Sector, NVIDIA
Responsible for global business development and strategy for the industrial sector at NVIDIA. Over the past 20 years, he has held the positions of CTO, Senior VP, and Head of Research Labs at various companies.
Dan Lesovodski
Partner, Data Monsters
Managed 46 AI projects for large companies, including GE, Intel, AB InBev, and Boston Scientific. Author of the course “AI for managers” on Coursera.
German Suvorov
Head of Industrial AI, Data Monsters
An expert in AI for industrial applications with over 20 years of experience in automotive manufacturing, advanced materials engineering, industrial automation, supply chain management, and information technologies.
PARTNERS
WHY CHOOSE US?
NetApp is a global leader in data management solutions, providing a common infrastructure for moving data from edge to core to cloud. They provide a simplified approach to managing data and accelerating applications whether on-premise, in the cloud, or both. Sycomp is a global system integrator, developing architectures that bring together data science, software and infrastructure to provide clients with a holistic approach to deploying AI/ML solutions. With their global footprint of over 40 countries, they bring a level of standardization to global technology deployments. Data Monsters. An NVIDIA Elite service delivery partner with a portfolio of 100+ AI projects.
