R4DGS: Referring Segmentation in 4D Gaussian Splatting
Computer Science > Computer Vision and Pattern Recognition arXiv:2610.30326 (cs) [Submitted on 21 Oct 2026] Title: R4DGS: Referring Segmentation in 4D Gaussian Splatting Authors: Bangpu Chen, Yaxuan Li, Shirui Peng, Xiangtian Si, Liuxin Chu, Xitong Cao, Hongbo Jin, Jiayu Ding.
计算机科学 > 计算机视觉与模式识别 arXiv:2610.30326 (cs) [提交于 2026 年 10 月 21 日] 标题：R4DGS：4D 高斯溅射中的指代分割 作者：Bangpu Chen, Yaxuan Li, Shirui Peng, Xiangtian Si, Liuxin Chu, Xitong Cao, Hongbo Jin, Jiayu Ding。
Abstract: Dynamic representations based on 4D Gaussian Splatting (4DGS) have shown strong capabilities in reconstruction and rendering, yet current methods remain limited in supporting complex natural-language referring in dynamic scenes. Existing 4D Gaussian methods are primarily designed for rendering-oriented dynamic modeling, while current semantic extensions often rely on scene-specific semantic optimization or are evaluated on relatively simple query settings.
摘要：基于 4D 高斯溅射（4DGS）的动态表示在重建和渲染方面展现出了强大的能力，但现有方法在支持动态场景中复杂的自然语言指代方面仍然受限。现有的 4D 高斯方法主要针对面向渲染的动态建模而设计，而当前的语义扩展往往依赖于特定场景的语义优化，或是在相对简单的查询设置下进行评估。
In this paper, we study Referring 4D Gaussian Splatting (R4DGS), a task for grounding temporally varying, multi-target or reasoning-intensive, and zero-target or distractor expressions in dynamic 4D scenes. To support this task, we introduce R4D-Bench-QA, a benchmark with structured query annotations. We further present ReferGaussian, a unified framework built upon 4DGS that converts low-level dynamic Gaussians into persistent, language-addressable entities for training-free referring inference.
在本文中，我们研究了指代 4D 高斯溅射（R4DGS），这是一项用于在动态 4D 场景中定位随时间变化的、多目标或推理密集型、以及零目标或干扰项表达式的任务。为了支持该任务，我们引入了 R4D-Bench-QA，这是一个带有结构化查询标注的基准测试。我们进一步提出了 ReferGaussian，这是一个建立在 4DGS 之上的统一框架，它将低级动态高斯转换为持久的、可语言寻址的实体，以实现无需训练的指代推理。
ReferGaussian organizes mask-supported Gaussian entities in an EntityBank, uses Multimodal Large Language Models (MLLMs) to decompose complex queries into trackable object phrases and structured referring constraints, and performs constraint-aware grounding without scene-specific semantic optimization. Experiments on R4D-Bench-QA and the 4D LangSplat protocol demonstrate improved temporal target selection and competitive spatiotemporal grounding without scene-specific semantic training.
ReferGaussian 在 EntityBank 中组织了支持掩码的高斯实体，利用多模态大语言模型（MLLMs）将复杂查询分解为可追踪的对象短语和结构化指代约束，并在无需特定场景语义优化的情况下执行约束感知定位。在 R4D-Bench-QA 和 4D LangSplat 协议上的实验表明，该方法在无需特定场景语义训练的情况下，改善了时间目标选择并实现了具有竞争力的时空定位。