arXiv:2608. 13255v1 Announce Type: cross Abstract: Geometry-conditioned multi-view diffusion enables high-quality 3D texture generation, but its repeated per-view denoiser evaluations introduce substantial computational cost.
By Haotang Li, Zhenyu Qi, Shaohan Henry Wang, Kebin Peng, Yutong Zhao, Zi Wang, Bo Liu, Huanrui Yang, Sen He
MeshSuite is the first benchmark designed to systematically evaluate triangle- and mesh-based neural rendering methods in both standard and dedicated engine deployment settings. It compares how different methods perform when using a conventional opaque mesh pipeline versus method-specific engine implementations, and also audits exported surfaces for topological and geometric integrity. The benchmark reveals that engine deployment generally degrades image quality, with MeshSplatting performing best under standard deployment, while dedicated deployment preserves fidelity at a higher computational cost.
By Kaixuan Zhang, Minxian Li, Mingwu Ren, Xiatian Zhu
T$^2$-Splat is a novel triangle‑based splatting method that separates topology adaptation from appearance refinement. It starts with Safe Restricted Delaunay Triangulation to set initial mesh connectivity, then uses Dynamic Topology Evolution to adjust triangle allocation via edge collapse and vertex split operations. Finally, residual per‑face Mip‑textures refine appearance on the fixed mesh with scale‑adaptive filtering, yielding competitive results on datasets such as Mip‑NeRF 360, Tanks & Temples, and DTU while allowing flexible mesh compression.
"whyItMatters":"The approach achieves higher PSNR (0.71 dB over MeshSplatting on Mip‑NeRF 360) and supports extreme mesh compression, enabling efficient novel view synthesis with compact meshes."
By Zhihao Tang, Youjia Zhang, Mingbo Zhao, Wei Yang
GraftSR is a diffusion-based super‑resolution framework that uses reference images of the same object to guide texture restoration, mitigating hallucination. It introduces a dual‑mask reference guidance mechanism to decouple texture extraction from application, avoiding reliance on spatial alignment. The authors also release TexRefSR‑141K, a large dataset of reference pairs with spatial masks, and show that GraftSR outperforms existing methods on the TexRefSR‑Eval benchmark, reducing LPIPS by 20.2%.
By Qifan Yu, Haoran Bai, Zongyao He, Weijie He, Sibin Deng, Honggang Qi, Ying Chen
arXiv:2607.13808v2 Announce Type: replace
Abstract: Neural radiance representations in Gaussian Splatting (GS) deliver high-fidelity color detail but impose substantial rendering overhead from networ...
By Neel Kelkar, Simon Niedermayr, Kaloian Petkov, Klaus Engel, R\"udiger Westermann
Splatting-based algorithms reconstruct photorealistic, real-time-renderable, and mesh-exportable 3D scenes from regular images, but they represent a scene as a single monolithic field. Therefore, the...