解决在线动态3D重建中时空不一致问题，提升静态区域稳定性。
Compensating Spatiotemporally Inconsistent Observations for Online Dynamic 3D Gaussian Splatting
- 通过学习并减去真实拍摄中的误差，恢复理想观测。
- 在多个数据集上显著提升时序一致性与渲染质量。
- 适合需要实时动态场景重建的科研与工业应用。
在线动态场景重建对从实时视频流中学习场景至关重要，而现有方法多依赖录制视频。然而，以往在线重建方法主要关注效率和渲染质量，忽视了结果的时间一致性，常导致静态区域出现明显伪影。本文指出，真实拍摄中的噪声等误差是造成在线重建时序不一致的根本原因。提出一种方法，通过学习并减去误差来恢复理想观测。实验表明，该方法应用于多个基线模型，在不同数据集上均显著提升了时序一致性和渲染质量。代码、视频结果及模型检查点已公开于 https://bbangsik13.github.io/OR2。
原文摘要 · Abstract (English)
Online reconstruction of dynamic scenes is significant as it enables learning scenes from live-streaming video inputs, while existing offline dynamic reconstruction methods rely on recorded video inputs. However, previous online reconstruction approaches have primarily focused on efficiency and rendering quality, overlooking the temporal consistency of their results, which often contain noticeable artifacts in static regions. This paper identifies that errors such as noise in real-world recordings affect temporal inconsistency in online reconstruction. We propose a method that enhances temporal consistency in online reconstruction from observations with temporal inconsistency which is inevitable in cameras. We show that our method restores the ideal observation by subtracting the learned error. We demonstrate that applying our method to various baselines significantly enhances both temporal consistency and rendering quality across datasets. Code, video results, and checkpoints are available at https://bbangsik13.github.io/OR2.
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。