EvGGS：基于事件的泛化高斯泼溅的协同学习框架
EvGGS: A Collaborative Learning Framework for Event-based Generalizable Gaussian Splatting
AI总结：
针对事件数据稀疏且无绝对颜色导致3D重建困难的问题，提出首个基于事件的泛化3D重建框架EvGGS，通过级联的深度估计、强度重建和高斯回归模块协同训练，实现无需重训即可泛化的高质量3D高斯重建。
摘要：
事件相机具有高动态范围和低延迟等显著优势，非常适合光照条件恶劣和快速运动的场景。然而，由于事件数据稀疏且不包含绝对颜色信息，从原始事件流重建3D场景十分困难。为释放其在3D重建中的潜力，我们提出首个基于事件的泛化3D重建框架EvGGS，该框架仅以前馈方式从事件输入中将场景重建为3D Gaussians，且无需重新训练即可泛化至未见场景。该框架包含深度估计模块、强度重建模块和高斯回归模块，这些子模块以级联方式连接，我们通过设计的联合损失进行协同训练，使其相互促进。为促进相关研究，我们构建了包含多种材质物体的新型基于事件的3D数据集，并提供灰度图、深度图、相机位姿和轮廓的校准标签。实验表明，联合训练的模型显著优于单独训练的模型。我们的方法在重建质量及深度/强度预测上优于所有基线方法，且渲染速度令人满意。
英文摘要：
Event cameras offer promising advantages such as high dynamic range and low latency, making them well-suited for challenging lighting conditions and fast-moving scenarios. However, reconstructing 3D scenes from raw event streams is difficult because event data is sparse and does not carry absolute color information. To release its potential in 3D reconstruction, we propose the first event-based generalizable 3D reconstruction framework, called EvGGS, which reconstructs scenes as 3D Gaussians from only event input in a feedforward manner and can generalize to unseen cases without any retraining. This framework includes a depth estimation module, an intensity reconstruction module, and a Gaussian regression module. These submodules connect in a cascading manner, and we collaboratively train them with a designed joint loss to make them mutually promote. To facilitate related studies, we build a novel event-based 3D dataset with various material objects and calibrated labels of grayscale images, depth maps, camera poses, and silhouettes. Experiments show models that have jointly trained significantly outperform those trained individually. Our approach performs better than all baselines in reconstruction quality, and depth/intensity predictions with satisfactory rendering speed.
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。
