文章预览
本期带来上周在arXiv公开的论文11篇,涉及3D、4D重建、3D生成、3D编辑。 一、3D、4D重建 1.[2024-arXiv] AniSDF: Fused-Granularity Neural Surfaces with Anisotropic Encoding for High-Fidelity 3D Reconstruction 作者机构:Jingnan Gao, et al. Shanghai Jiao Tong University 论文地址: https://arxiv.org/pdf/2410.01202 项目地址: https://g-1nonly.github.io/AniSDF_Website/ 贡献: We design a unified SDF-based architecture that the geometry network and the appearance network complement each other, producing high-fidelity 3D reconstructions. We present a fused-granularity neural surface to balance the overall structures and fine details. We introduce blended radiance fields with a physics-based rendering via Anisotropic Spherical Gaussian encoding, successfully disambiguating the reflective appearance. Oue method boosts the quality of SDF-based methods by a great scale in both geometry reconstruction and novel-view synthesis tasks. 2.[2024-arXiv] Space-ti
………………………………