What problem does it solve?
This Skill helps create persistent, downloadable 3D environments from text, images, video, or 360 panoramas when a flat image or short video is not enough. It guides users through model selection, spatial control, export decisions, quality review, and licensing so generated worlds are suitable for VR, web experiences, virtual production, and game workflows.
Core Features & Use Cases
- Multimodal World Generation: Create bounded, navigable 3D environments from text, single or multiple images, video, and equirectangular panoramas.
- Spatial and Production Control: Use Chisel for room layouts and structural blocking, select between draft, standard, and expanding Marble models, and apply editing or expansion workflows.
- Downstream Delivery: Choose Gaussian splats for visual fidelity, collider meshes for physics, or high-quality meshes for editable geometry across web, Unity, Unreal, Blender, and Houdini.
- Use Case: Turn a studio photograph or panorama into an explorable virtual set, review seams and occluded regions, then export the appropriate assets for a commercial VR or game project with rights and input provenance checked.
Quick Start
Use the world-labs-marble skill to turn the provided studio image into a reviewed explorable 3D environment, recommending the best input mode, model, export format, downstream renderer, and licensing plan.
