提出概率流蒸馏，实现3D生成的精确分布匹配。
Probability-Flow Distillation: Exact Wasserstein Gradient Flow for High-Fidelity 3D Generation
- 用概率流替代后验均值估计，避免梯度近似误差。
- 生成结果更精细，显著减少过平滑与过饱和问题。
- 适合追求高保真3D内容生成的研究者与开发者。
Score Distillation Sampling（SDS）及其变体通过蒸馏2D图像扩散先验广泛用于文本到3D生成。然而，标准SDS目标易导致严重模式崩溃，常产生过平滑、过饱和的结果。尽管近期方法如通过反演实现的分数蒸馏（SDI）缓解了这些伪影并生成更清晰的模型，但最终仍无法忠实捕捉完整目标分布。本文揭示，限制SDI采样能力的根本原因在于其依赖后验均值估计，这在数学上等价于确定性反向DDIM轨迹的单步欧拉近似。为此，我们提出自然延伸方法——概率流蒸馏（PFD）。我们证明PFD恰好对应于Wasserstein梯度流，从而诱导出原则性的分布匹配动态。最后，实验表明PFD可生成具有细粒度、高保真细节的3D资产，且质量优于现有方法。
原文摘要 · Abstract (English)
Score Distillation Sampling (SDS) and its variants have been widely used for text-to-3D generation by distilling 2D image diffusion priors. However, the standard SDS objective is prone to severe mode collapse, frequently yielding over-smoothed and over-saturated results. Although recent advancements, such as Score Distillation via Inversion (SDI), mitigate these artifacts and produce visually sharper models, they ultimately fail to faithfully capture the full target distribution. In this work, we show that the bottleneck limiting the sampling capacity of SDI stems from its reliance on the posterior mean estimator, which is mathematically equivalent to a single-step Euler approximation of the deterministic reverse DDIM trajectory. To address this, we propose a naturally motivated extension termed Probability-Flow Distillation (PFD). We establish that PFD corresponds exactly to a Wasserstein gradient flow, thereby inducing principled distribution-matching dynamics. Finally, we show that PFD can synthesize 3D assets with fine-grained, high-fidelity details and achieve improved quality compared to existing methods.
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。