Understanding Zip-NeRF - a cool new AI algorithm for 3D scene synthesis | Neural Breakdown

In this video, I discuss the paper Zip-NeRF: Anti-Aliased Grid-Based Neural Radiance Fields by Barron et. al. which is a technique that achieves some amazing results for synthesizing 3D scene representations from 2D images. I describe the different components that make this technology possible with an emphasis on previous techniques like NeRF, MipNeRF, Instant NGP, and how Zip-NeRF improves on these methods! A lot of the footage is taken from existing resources. Full credit to them, here are some important links: NeRF project page - https://www.matthewtancik.com/nerf MipNeRF project page - https://jonbarron.info/mipnerf/ MipNeRF-360 project page - https://jonbarron.info/mipnerf360/ Instant NGP project page - https://nvlabs.github.io/instant-ngp/ ZipNeRF project page - https://jonbarron.info/zipnerf/ ZipNeRF arxiv link: https://arxiv.org/abs/2304.06706