PhysForge: Generating Physics-Grounded 3D Assets for Interactive Virtual World

Synthesizing physics-grounded 3D assets is a critical bottleneck for interactive virtual worlds and embodied AI. Existing methods predominantly focus on static geometry, overlooking the functional properties essential for interaction. We propose that interactive asset generation must be rooted in functional logic and hierarchical physics. To bridge this gap, we introduce PhysForge, a decoupled two-stage framework supported by PhysDB, a large-scale dataset of 150,000 assets with four-tier physical annotations. First, a VLM acts as a physical architect to plan a Hierarchical Physical Blueprint defining material, functional, and kinematic constraints. Second, a physics-grounded diffusion model realizes this blueprint by synthesizing high-fidelity geometry alongside precise kinematic parameters via a novel KineVoxel Injection (KVI) mechanism. Experiments demonstrate that PhysForge produces functionally plausible, simulation-ready assets, providing a robust data engine for interactive 3D content and embodied agents.

Virtual worlds, games, and robot simulators need 3D objects that not only look realistic, but can also be interacted with in realistic ways. Current 3D generation methods mostly create static shapes, so the generated objects often cannot be opened, pushed, grasped, or moved properly. This paper introduces PhysForge, a system that creates 3D objects from a single image while also describing their parts, materials, functions, and movable components. For example, it can identify that a cabinet has doors and handles, and generate how those parts should move. The resulting assets can be used more directly in robot simulation, games, and interactive virtual environments.