Wild3R: 从无约束稀疏照片集合进行前馈式3D高斯泼溅
Wild3R: Feed-Forward 3D Gaussian Splatting from Unconstrained Sparse Photo Collection
- The University of Tokyo（东京大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
提出Wild3R，一种针对无约束稀疏照片集合的前馈式3D高斯泼溅方法，通过引入包含多样光照和瞬态物体的WildCity数据集，学习跨视角外观一致性并移除瞬态内容，性能优于现有前馈方法，与基于逐场景优化的方法相当。
AI中文摘要：
前馈式3D高斯泼溅（3DGS）消除了传统3DGS所需的耗时逐场景优化。然而，现有的前馈方法难以处理包含多样光照条件和瞬态物体的真实世界照片集合。在本文中，我们提出了Wild3R，一种针对无约束稀疏照片集合的前馈方法。主要瓶颈在于缺乏提供多视角、多种光照和瞬态变化的训练数据，而这些是学习鲁棒场景表示所必需的。为解决这一问题，我们引入了WildCity数据集，该数据集包含200个场景、170种光照条件和瞬态物体，总计337,500张图像。通过利用该数据集，我们的模型在参考视图条件下学习跨视角的外观一致性，同时移除瞬态内容。大量实验表明，我们的方法优于现有的前馈方法，并取得了与先前基于逐场景优化的方法相竞争的结果。
英文摘要：
Feed-forward 3D Gaussian Splatting (3DGS) removes the need for time-consuming per-scene optimization required by traditional 3DGS. However, existing feed-forward approaches struggle with real-world photo collections that include diverse lighting conditions and transient objects. In this paper, we present Wild3R, a feed-forward approach for unconstrained sparse photo collections. The main bottleneck is the lack of training data that provides multiple viewpoints, a variety of illuminations, and transient variations necessary for learning robust scene representations. To address this, we introduce the WildCity dataset, which comprises 200 scenes, 170 lighting conditions, and transient objects, resulting in 337,500 images in total. By leveraging the dataset, our model learns appearance consistency across viewpoints conditioned on reference views, while removing transient content. Extensive experiments demonstrate that our method outperforms existing feed-forward approaches and achieves results competitive with prior per-scene optimization-based methods.
