Neural Fields for Scalable Scene Reconstruction

From Design Computation
Jump to: navigation, search
DCIO2022-Logo.png
DC I/O 2022 Keynote by James Tompkin. https://doi.org/10.47330/DCIO.2022.AXBL8798 | Left | Left | Left | Left


DCIO2022 S1 1 J-Tompkin.png


Abstract

Neural fields are a new (and old!) approach to solving problems over spacetime via first-order optimization of a neural network. Over the past three years, combining neural fields with classic computer graphics approaches have allowed us to make significant advances in solving computer vision problems like scene reconstruction. I will present recent work that can reconstruct indoor scenes for photorealistic interactive exploration using new scalable hybrid neural field representations. This has applications where any real-world place needs to be digitized, especially for visualization purposes.

Keywords

AI, Architect, Neural Fields.

Bibliography

  • Anderson, T.T., 2011. Complicating Heidegger and the Truth of Architecture. The Journal of Aesthetics and Art Criticism 69, 69–79.