Seonglae ChoScore Distillation Sampling
First introduced in Google's DreamFusion
Score Distillation Sampling (SDS Loss)
A new loss function that extracts gradients from the diffusion model's score function to optimize randomly initialized NeRF parameters.
Score predictions (denoising direction) and actual noisy data differences are used to calculate MSE gradient (parameter direction), aiming for a density distillation loss effect
[논문리뷰] DreamFusion: Text-to-3D using 2D Diffusion
2D diffusion model을 3D object synthesis에서 사용가능하게 하는 Score Distillation Sampling(SDS)를 소개한다
![[논문리뷰] DreamFusion: Text-to-3D using 2D Diffusion](https://jang-inspiration.com/favicon.png){kind=small}
https://jang-inspiration.com/dreamfusion