# NEO：一次 NeRF，多次编辑，用于连续物体操作

> 原文链接：https://robotics.hellobike.com/papers/2026-07-27-2607-24538
> 频道：papers | 更新日期：2026-07-27 | 标签：arxiv, cs.RO
> 核心摘要：NEO面向连续物体操作，用语言引导NeRF编辑，并在场景编辑、物体移除和pick-and-place实验中验证。

# NEO：一次 NeRF，多次编辑，用于连续物体操作

::: v-pre
> 原文链接：[NEO: NeRF It Once, Edit It Many Times for Continuous Object Manipulation](https://arxiv.org/abs/2607.24538)（arXiv 2607.24538 · 发布日期：2026-07-27）

**NEO: NeRF It Once, Edit It Many Times for Continuous Object Manipulation**

NEO面向连续物体操作，用语言引导NeRF编辑，并在场景编辑、物体移除和pick-and-place实验中验证。

## 核心要点

- 针对机器人操作中的NeRF场景编辑，目标是在动作执行前连贯建模未来场景状态。
- 提出语言引导物体移除，结合神经场重采样与多视图一致的渐进式修复。
- 通过知识蒸馏直接编辑NeRF权重，以教师-学生模型组合原始与编辑NeRF。
- 构建NEO-Dataset用于定量评估；方法优于现有基线，编辑更视觉连贯、几何一致。

**验证范围与局限**：摘要仅说明在场景编辑、物体移除和pick-and-place机器人实验中验证。

## 英文摘要 (Abstract)

In this paper, we present NEO, a unified framework providing language-guided NeRF editing for robotic manipulation. Our paper introduces (i) a language-guided object removal that combines neural field resampling with multiview-consistent progressive inpainting, (ii) a direct NeRF weight editing method utilizing knowledge distillation, composing original and edited NeRFs via a teacher-student model, enabling coherent modeling of future scene states before a robot executes an action, and (iii) the first benchmark (NEO-Dataset) for quantitatively evaluating NeRF scene editing methods suitable for robot manipulation. We show that our approach outperforms state-of-the-art baselines in scene editing tasks, including object removal and pick-and-place robotic experiments, yielding visually coherent and geometrically consistent edits that reduce artifacts commonly introduced by prior methods.

作者：Mikołaj Zieliński, David Hall, Dominik Belter, Peyman Moghadam
:::

## 原始参考来源 (Sources)

- [arXiv 2607.24538：NEO: NeRF It Once, Edit It Many Times for Continuous Object Manipulation](https://arxiv.org/abs/2607.24538)