Zhenjun Zhao (@zhenjun_zhao)
2025-08-14 | โค๏ธ 50 | ๐ 6
GSFixer: Improving 3D Gaussian Splatting with Reference-Guided Video Diffusion Priors
Xingyilang Yin, Qi Zhang, et al.
tl;dr: 2D semantic (from DINOv2)+3D geometric (from VGGT) tokens from input viewsโvideo diffusion modelโartifact-prone novel views; reference trajectory samplingโangular coverage and restoration quality
https://arxiv.org/abs/2508.09667
๋ฏธ๋์ด
