Matthias Niessner (@MattNiessner)

2024-10-18 | โค๏ธ 317 | ๐Ÿ” 58


๐Ÿ“ข๐‹๐Ÿ‘๐ƒ๐†: ๐‹๐š๐ญ๐ž๐ง๐ญ ๐Ÿ‘๐ƒ ๐†๐š๐ฎ๐ฌ๐ฌ๐ข๐š๐ง ๐ƒ๐ข๐Ÿ๐Ÿ๐ฎ๐ฌ๐ข๐จ๐ง๐Ÿ“ข SIGGRAPHAsia

We propose a generative diffusion model for 3D Gaussians. Key is a learnt latent space which substantially reduces the complexity of the diffusion process, thus facilitating room-scale scene generation!

https://www.youtube.com/watch?v=UHEEiXCYeLU&feature=youtu.be https://barbararoessle.github.io/l3dg/

Great work by @barbara_roessle in with @normanisation, @angelaqdai, Lorenzo Porzi, Samuel Rota Bulรฒ, Peter Kontschieder!

๋ฏธ๋””์–ด

video


Tags

domain-vision-3d domain-genai