Zexiang Xu (@zexiangxu)
2024-10-24 | โค๏ธ 89 | ๐ 16
No NeRF~ No 3DGS~ Our LVSM delivers state-of-the-art view synthesis quality from sparse-view inputs! Time to fully embrace scalable large transformers for building โ3D-freeโ 3D models and apps!
Amazing work by our intern @Haian_Jin and collaborators. ๐https://haian-jin.github.io/projects/LVSM/
์ธ์ฉ
[@Haian_Jin]: Novel view synthesis has long been a core challenge in 3D vision. But how much 3D inductive bias is truly needed? โSurprisingly, very little!
Introducing โLVSM: A Large View Synthesis Model with Miniโฆ
์ธ์ฉ ํธ์
Haian Jin (@Haian_Jin)
Novel view synthesis has long been a core challenge in 3D vision. But how much 3D inductive bias is truly needed? โSurprisingly, very little!
Introducing โLVSM: A Large View Synthesis Model with Minimal 3D Inductive Biasโโa fully transformer-based approach that enables scalable, generalizable, and fully data-driven novel view synthesis, from sparse posed inputs. ๐งต(1/6)
Project Page: https://t.co/Alqo3s0Qjt
๐ฌ ์์
