BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Creation of Virtual Worlds for Machine Learning
DTSTART:20260922T070000Z
DTEND:20260923T140000Z
DTSTAMP:20260901T052400Z
UID:indico-event-508@events.hpc-portal.eu
CONTACT:t.george@fz-juelich.de
DESCRIPTION:This course will take place as an online event. The link to th
 e streaming platform will be provided to the registrants only. The course 
 will be held in English. \nCourse Content:\nUnreal Engine is a powerful 3
 D rendering tool widely used in game development. In recent years\, it has
  become increasingly popular in science and industry. This course teaches 
 you how to use Unreal Engine to generate synthetic data for machine learni
 ng projects.\nYou will learn to create 3D assets using AI tools (ComfyUI)\
 , build large virtual worlds automatically using Unreal's Procedural Conte
 nt Generation (PCG) framework\, export video footage from these worlds via
  Network Device Interface (NDI)\, and prepare the data for training video 
 generation models. By the end of the course\, you will have built a comple
 te working pipeline and understand how to use it for machine learning work
 flows.\nWhat you will learn:\n\nGenerate 3D assets using AI (ComfyUI) and 
 import them into Unreal Engine\nBuild scalable virtual worlds with Unreal'
 s PCG framework\nExport camera footage and metadata from virtual scenes us
 ing NDI (network video over IP)\nGetting to know city sample and how NDI o
 utputs data streams for training data\nOrganise and prepare synthetic data
 sets for machine learning\nUnderstand training pipelines for video generat
 ion models\n\nPrerequisites:\nBasic familiarity with 3D/graphics concepts.
  Understanding of meshes\, materials\, or 3D modeling is helpful but not r
 equired. Basic Python or scripting experience is useful for data pipeline 
 automation\, but not essential.\nFamiliarity with machine learning concept
 s is recommended such as a general understanding of datasets\, train/valid
 ation splits\, and model inference. We assume participants are familiar wi
 th general concepts of machine learning and deep learning. For an introduc
 tion to these topics\, we refer to open resources:\nPrevious knowledge of 
 Unreal Engine is essential\, at least knowledge of viewport and blueprints
 .\nTo be well prepared\, install in advance the Unreal Engine 5.1.* and a 
 C++ Development IDE such as Visual Studio 2022\, as well as ComfyUI with T
 rellis support. Details are available on https://gitlab.jsc.fz-juelich.de
 /hedgedoc/24tF29QsTe-ijNBHU1I5VQ?both  where we are updating information
  as problems and questions arise.\nA personal institutional email address 
 (university/research institution\, government agency\, organisation\, or c
 ompany) is required to register for JSC training courses. If you don't hav
 e an institutional email address\, please get in touch with the contact pe
 rson for this course.\nTarget Audience:\n\nScientists and engineers buildi
 ng synthetic datasets for computer vision\, 3D reconstruction\, or video g
 eneration research\nDevelopers of analysis pipelines that consume syntheti
 c camera data or structured 3D scene parameters\nResearchers interested in
  controlled\, reproducible virtual scene generation for model evaluation o
 r training\nEngineers deploying Unreal Engine as a visualization or data-g
 eneration tool in scientific and industrial settings\nAnyone interested in
  understanding the complete pipeline from 3D scene creation to machine lea
 rning model training setup\n\nLanguage:\nThis course is taught in English.
 \nDuration:\n2 days\nDates:\n22-23 September 2026\, 09:00-12:00\, 13:00-16
 :00 each day\nVenue:\nOnline via Zoom\nNumber of Participants:\nMinimum 5\
 , maximum 30\nInstructor:\nThomas George\, JSC\n\nhttps://events.hpc-porta
 l.eu/event/508/
LOCATION:Online
URL:https://events.hpc-portal.eu/event/508/
END:VEVENT
END:VCALENDAR