GSComplete:基于2D扩散先验的高斯泼溅补全
GSComplete: Gaussian Splat Completion with 2D Diffusion Priors
浏览论文内容
中文总结 AI 辅助
GSComplete提出结合分数蒸馏采样与保留损失,仅用2D扩散先验补全高斯泼溅对象,保留原始泼溅并仅生成缺失区域,新数据集上验证了更优的输入保留性能。
中文摘要 AI 辅助
高斯泼溅为3D对象提供了一种快速、高保真的表示,但在实践中,它们通常由不完整的输入数据构建,从而留下缺失区域。现有的补全方法要么不保留原始泼溅,要么需要稀缺的3D训练数据。我们提出了GSComplete,它将基于分数蒸馏采样的3D生成与一种新颖的保留损失相结合,该损失鼓励在原始泼溅应可见的位置对其进行保留。这有效地仅使用2D扩散先验完成了高斯泼溅对象的补全,同时完全保留现有泼溅,并仅在缺失区域生成新泼溅,而不会遮挡输入。为了评估我们的方法,我们引入了一个新的部分高斯泼溅对象数据集,并表明GSComplete在保留输入方面比现有方法显著更准确,同时补全结果的合理性相当。我们的代码和数据集将在录用后公开。
英文摘要
Gaussian splats provide a fast, high-fidelity representation for 3D objects but are often constructed from incomplete input data in practice, leaving missing regions. Existing completion methods either do not preserve the original splats or require scarcely available 3D training data. We propose GSComplete, which combines 3D generation based on Score Distillation Sampling with a novel preservation loss that encourages the original splats to be preserved where they should be visible. This effectively completes the Gaussian splat object using only 2D diffusion priors while fully preserving existing splats and generating new splats only in missing regions, without occluding the input. To evaluate our approach, we introduce a new dataset of partial Gaussian splat objects and show that GSComplete achieves significantly more accurate preservation of the input than existing methods with comparable plausibility of the completed result. Our code and dataset will be made available upon acceptance.
发表机构
- TU Wien(维也纳工业大学)
- Adobe Research(奥多比研究院)
机构由 AI 辅助整理,请以论文原文为准。