发表机构
University of Bologna; Peking University; The Hong Kong Polytechnic University; Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ); Huawei; University of Trento; University of the Chinese Academy of Sciences; Monash University; Google DeepMind; University of California, Merced（博洛尼亚大学; 北京大学; 香港理工大学; 广东省人工智能与数字经济实验室（深圳）; 华为; 特伦托大学; 中国科学院大学; 莫纳什大学; 谷歌DeepMind; 加州大学默塞德分校）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出4D场景重建的统一视角，围绕场景表示、时间建模、重建流程和优化目标组织现有方法，分析设计选择对几何、外观、运动和效率的影响，并整合数据集与评估指标，为未来研究提供结构化基础。
AI 中文摘要
4D场景重建旨在从视觉观测中恢复动态环境的演化几何、外观和运动。尽管神经场景表示取得了显著进展，但由于非刚性运动、遮挡、时间不一致性以及重建保真度与计算效率之间的权衡，重建动态场景仍然具有挑战性。神经辐射场（NeRF）和3D高斯泼溅（3DGS）的最新进展引入了多种表示和重建动态场景的方法，然而它们之间的关系、底层设计选择以及评估协议仍然零散。在本文中，我们提出了一个关于4D场景重建的统一视角，围绕场景表示、时间建模策略、重建流程和优化目标来组织现有方法。通过这一框架，我们考察了不同设计选择如何影响几何保真度、外观一致性、运动表示和计算效率。我们进一步整合了常用的数据集和评估指标，识别了当前实验实践中的局限性，并讨论了重建复杂动态真实世界环境中的开放挑战。通过将方法论进展与其底层假设和评估证据联系起来，本工作为理解现有方法和识别未来研究方向提供了结构化基础。相关论文和资源的持续更新集合可在该https URL获取。
英文摘要
4D scene reconstruction aims to recover the evolving geometry, appearance, and motion of dynamic environments from visual observations. Despite substantial progress in neural scene representations, reconstructing dynamic scenes remains challenging due to non-rigid motion, occlusions, temporal inconsistencies, and the trade-offs between reconstruction fidelity and computational efficiency. Recent advances in Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS) have introduced diverse approaches to representing and reconstructing dynamic scenes, yet their relationships, underlying design choices, and evaluation protocols remain fragmented. In this paper, we present a unified perspective on 4D scene reconstruction, organizing existing methods around their scene representations, temporal modeling strategies, reconstruction pipelines, and optimization objectives. Through this framework, we examine how different design choices affect geometric fidelity, appearance consistency, motion representation, and computational efficiency. We further consolidate commonly used datasets and evaluation metrics, identify limitations in current experimental practices, and discuss open challenges in reconstructing complex, dynamic real-world environments. By connecting methodological developments with their underlying assumptions and evaluation evidence, this work provides a structured foundation for understanding existing approaches and identifying future research directions. An evolving collection of relevant papers and resources is available at https://github.com/ZiyangYan/Awesome-4D-Scene-Reconstruction.
