用稀疏体素提升3D生成纹理质量，解决细节丢失与内存占用难题
Sparse-Up: Learnable Sparse Upsampling for 3D Generation with High-Fidelity Textures
- 通过可学习的稀疏上采样，将体素约束在网格表面，减少超70%冗余体素
- 采用视域分区策略，仅对可见局部图像块反向传播梯度，降低内存消耗
- 在保持几何一致性的同时，显著提升纹理高频细节，适合高保真3D内容生成
高质量3D资产生成常受限于‘像素级痛点’：高频细节丢失。现有方法往往在跨视角一致性与分辨率之间权衡——或导致纹理撕裂、漂移，或受限于显式体素的分辨率上限，牺牲精细纹理。本文提出Sparse-Up，一种高效内存、高保真纹理建模框架。通过稀疏体素引导纹理重建，结合表面锚定与视域分区，突破分辨率瓶颈。表面锚定采用可学习上采样策略，将体素约束在网格表面，消除超过70%传统上采样中的冗余体素。视域分区引入图像块引导的体素分区机制，仅对可见局部块进行监督与梯度回传。该策略在高分辨率体素训练中显著降低内存开销，同时保持几何一致性，并完整保留纹理高频细节。
原文摘要 · Abstract (English)
The creation of high-fidelity 3D assets is often hindered by a 'pixel-level pain point': the loss of high-frequency details. Existing methods often trade off one aspect for another: either sacrificing cross-view consistency, resulting in torn or drifting textures, or remaining trapped by the resolution ceiling of explicit voxels, forfeiting fine texture detail. In this work, we propose Sparse-Up, a memory-efficient, high-fidelity texture modeling framework that effectively preserves high-frequency details. We use sparse voxels to guide texture reconstruction and ensure multi-view consistency, while leveraging surface anchoring and view-domain partitioning to break through resolution constraints. Surface anchoring employs a learnable upsampling strategy to constrain voxels to the mesh surface, eliminating over 70% of redundant voxels present in traditional voxel upsampling. View-domain partitioning introduces an image patch-guided voxel partitioning scheme, supervising and back-propagating gradients only on visible local patches. Through these two strategies, we can significantly reduce memory consumption during high-resolution voxel training without sacrificing geometric consistency, while preserving high-frequency details in textures.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。