Gaussian Splatting: A New Neural Imagination Engine for 3D Data Representation

Explore the revolutionary world of Gaussian splatting in this 37-minute talk by Oles Petriv, CTO and co-founder of Reface. Delve into how this cutting-edge neural imagination engine is reshaping visual and semantic representation, reducing machine learning training costs, and seamlessly integrating into 3D software, game engines, and existing pipelines. Gain valuable insights into the future of artificial intelligence and real-time applications as Petriv, a Ukrainian AI pioneer with over a decade of expertise in computer vision and natural language processing, guides you through various 3D data representations, neural radiance fields, and the intricacies of Gaussian splatting. Learn about 4D Gaussian splatting, InstantSplat, and Long Splat, and discover how these innovations are transforming the fields of data science, AI, and machine learning.

- Introduction
- What types of 3D data representations Exist?
- What are neural radiance fields?
- What about time?
- What is Gaussian Splatting
- 4D Gaussian Splatting
- InstantSplat
- Long Splat