LAMP：将图像编辑提升为通用3D先验以实现开放世界操控
LAMP: Lift Image-Editing as General 3D Priors for Open-world Manipulation
AI总结：
LAMP通过将图像编辑作为3D先验，提取物体间连续的几何表示，提升开放世界操控的泛化能力与精度。
摘要：
人类般的开放世界泛化仍是机器人操控的基本挑战。现有基于学习的方法，包括强化学习、模仿学习和视觉-语言-动作模型（VLAs），在面对新任务和未见环境时往往表现不佳。另一个有前途的方向是探索能够捕捉细粒度空间和几何关系的可泛化表示。尽管大语言模型（LLMs）和视觉-语言模型（VLMs）基于语言或标注的2D表示提供了强大的语义推理能力，但其有限的3D感知限制了其在细粒度操控中的应用。为此，我们提出了LAMP，将图像编辑提升为3D先验，以提取物体间的3D变换作为连续、几何感知的表示。我们的核心见解是图像编辑本质上编码了丰富的2D空间线索，将这些隐含线索提升为3D变换，为开放世界操控提供精细且准确的指导。大量实验表明，\codename能够提供精确的3D变换，并在开放世界操控中实现强大的零样本泛化能力。项目页面：https://zju3dv.github.io/LAMP/。
英文摘要：
Human-like generalization in open-world remains a fundamental challenge for robotic manipulation. Existing learning-based methods, including reinforcement learning, imitation learning, and vision-language-action-models (VLAs), often struggle with novel tasks and unseen environments. Another promising direction is to explore generalizable representations that capture fine-grained spatial and geometric relations for open-world manipulation. While large-language-model (LLMs) and vision-language-model (VLMs) provide strong semantic reasoning based on language or annotated 2D representations, their limited 3D awareness restricts their applicability to fine-grained manipulation. To address this, we propose LAMP, which lifts image-editing as 3D priors to extract inter-object 3D transformations as continuous, geometry-aware representations. Our key insight is that image-editing inherently encodes rich 2D spatial cues, and lifting these implicit cues into 3D transformations provides fine-grained and accurate guidance for open-world manipulation. Extensive experiments demonstrate that \codename delivers precise 3D transformations and achieves strong zero-shot generalization in open-world manipulation. Project page: https://zju3dv.github.io/LAMP/.
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。
