LAMP: Lift Image-Editing as General 3D Priors for Open-world Manipulation

arXiv:2604.08475v1 Announce Type: new Abstract: 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 odename delivers precise 3D transformations and achieves strong zero-shot generalization in open-world manipulation. Project page: https://zju3dv.github.io/LAMP/.

Related

• DailyArt: Discovering Articulation from Single Static Images via Latent Dynamics
• GEAR: GEometry-motion Alternating Refinement for Articulated Object Modeling with Gaussian Splatting
• ReconPhys: Reconstruct Appearance and Physical Attributes from Single Video
• Uni-ViGU: Towards Unified Video Generation and Understanding via A Diffusion-Based Video Generator

Source: arXiv cs.CV | 2026-04-10