ABot-3DWorld 0: A Universal World Model to Explore Any 3D Space

2026-07-13

AI Summary

This work proposes a general-purpose multimodal 3D world generation framework that efficiently constructs high-fidelity, explorable 3D scenes from arbitrary text, image, or video inputs. The key innovation lies in a unified Spatial Generation Primitive (SGP) representation, which maps diverse modalities into compact geometric and appearance priors. By integrating 3D-consistent omnidirectional video synthesis with trajectory planning, the method enables coherent scene layout inference, followed by high-quality reconstruction via 3D Gaussian Splatting. The approach unifies processing across sparse to dense inputs, achieving state-of-the-art performance among open-source solutions. It significantly improves scene fidelity over Marble while supporting large-scale interactive exploration on consumer-grade hardware.

Abstract

We present ABot-3DWorld 0, a universal multimodal 3D world model that turns text, image, and video inputs into high-fidelity, explorable 3D worlds. At the heart of our framework is a unified Spatial Generative Primitive (SGP), a compact tuple of a high-quality panorama and a spatial point cloud that delivers an efficient description of any 3D space. Multimodal inputs are first lifted into this primitive; a 3D-consistent panoramic video generator then explores the primitive along a planned trajectory; finally, our panoramic video reconstruction engine converts the generated video into a clean, photorealistic 3D Gaussian Splatting (3DGS) world. This pipeline covers two regimes: rich inputs (multi-view sets, casual video) are lifted into the SGP through a geometry-rigorous recovery that mirrors the observed scene, while a single image or sentence is completed generatively into a creative world. The result is one low-barrier engine for general 3D content creation that further anchors generated worlds to geographic points of interest, enabling map-native spatial exploration at consumer scale. Experiments show that ABot-3DWorld 0 sets the state of the art among open-source methods and demonstrates stronger scene fidelity than Marble under rich multimodal inputs.

Problem

• 3D world model
• multimodal input
• spatial exploration
• 3D content creation
• photorealistic reconstruction

Innovation

• Spatial Generative Primitive
• 3D Gaussian Splatting
• multimodal 3D generation
• panoramic video reconstruction
• universal world model