The advent of photorealistic scene representations such as Neural Radiance Fields (NeRF) and 3D Gaussian Splattings (3DGS) enable us to replicate complex real-world elements. This paradigm shift in graphics lets us revisit the way of capturing, storing, and replicating the real world and rethink what to bring in XR (X for any type).

This tutorial provides an opportunity for learning the basics to "start with" future XR research around the topic. Namely, the attendees will learn the principles of view synthesis using machine learning technology and practical aspects of implementations using open-source projects. It also provides discussions and networking with VR attendees who share the same interests.

The tutorial will begin with a general overview, summarizing the latest developments and current trends in radiance field research for XR applications. This will be followed by an in-depth, live demonstration and tutorial on creating basic NeRF and 3DGS scenes for VR using readily available open-source software such as immersive-ngp and UnityGaussianSplatting toolkit. To conclude, we will summarize the key takeaways and facilitate an interactive discussion with the audience, exploring future trends and developments in radiance fields for XR.