Nix 和 Fix：通过扩散模型实现 3D 高斯散射的 1000 倍压缩
Nix and Fix: Targeting 1000x Compression of 3D Gaussian Splatting with Diffusion Models
AI总结：
本文提出 NiFi 方法，通过扩散模型实现 3DGS 在极低速率下的高质量压缩，达到 1000 倍压缩率。
摘要：
3D 高斯散射（3DGS）革新了视图渲染。与隐式表示不同，3DGS 使用稀疏高斯，这使实时性能得以实现，但增加了空间需求，阻碍了受速率限制的应用。3DGS 压缩成为缓解此问题的领域。尽管取得了显著进展，但在低速率下，压缩会引入伪影，显著降低视觉质量。我们引入 NiFi，一种通过基于扩散的、伪影感知的一步蒸馏进行极端 3DGS 压缩的方法。我们证明，我们的方法在极低速率（低至 0.1 MB）下实现了最先进的感知质量，并在与 3DGS 相当的感知性能下实现了 1000 倍的压缩率。代码可在：https://github.com/ceteke/nifi 获得。
英文摘要：
3D Gaussian Splatting (3DGS) revolutionized novel view rendering. Instead of inferring from dense spatial points, as implicit representations do, 3DGS uses sparse Gaussians. This enables real-time performance but increases space requirements, hindering rate-constrained applications. 3DGS compression emerged as a field aimed at alleviating this issue. While impressive progress has been made, at low rates, compression introduces artifacts that degrade visual quality significantly. We introduce NiFi, a method for extreme 3DGS compression through restoration via artifact-aware, diffusion-based one-step distillation. We show that our method achieves state-of-the-art perceptual quality at extremely low rates, down to 0.1 MB, and towards 1000x rate improvement over 3DGS at comparable perceptual performance. Code is available at: https://github.com/ceteke/nifi
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
