Learning Implicit Body Representations from Double Diffusion Based Neural Radiance Fields — mentioned by zone411

Thu, 01 Sep 2022 18:02:38 GMT

> "Stable Diffusion was trained primarily on LAION-Aesthetics, a collection of subsets of the LAION 5B dataset, which is itself a subset of LAION-400M" Err, no? As you can tell by the name itself, the 5B dataset is larger. > "Unlike autoencoder-based deepfake content, or the human recreations that can be achieved by Neural Radiance Fields (NeRF) and Generative Adversarial Networks (GANs), diffusion-based systems learn to generate images by adding noise" This is confused. Diffusion is orthogonal to NeRF. For example, here is a paper that uses both: https://arxiv.org/abs/2112.12390 > "Within…

Paper and discussion history. A mention is not an endorsement.

Learning Implicit Body Representations from Double Diffusion Based Neural Radiance Fields · HackerPapers discussions

Learning Implicit Body Representations from Double Diffusion Based Neural Radiance Fields — mentioned by zone411