Matrix-3D: Omnidirectional Explorable 3D World Generation

TL;DR

Matrix-3D introduces a panoramic-based framework for wide-coverage, explorable 3D world generation from a single image or text, combining novel diffusion models and reconstruction methods to improve scene quality and geometric consistency.

Contribution

The paper presents a new panoramic video diffusion model and two 3D reconstruction methods, along with a large-scale synthetic dataset, advancing 3D scene generation from limited inputs.

Findings

Achieves state-of-the-art panoramic video generation quality

Enables high-quality, geometrically consistent 3D scene reconstruction

Demonstrates effective 3D world generation from minimal input data

Abstract

Explorable 3D world generation from a single image or text prompt forms a cornerstone of spatial intelligence. Recent works utilize video model to achieve wide-scope and generalizable 3D world generation. However, existing approaches often suffer from a limited scope in the generated scenes. In this work, we propose Matrix-3D, a framework that utilize panoramic representation for wide-coverage omnidirectional explorable 3D world generation that combines conditional video generation and panoramic 3D reconstruction. We first train a trajectory-guided panoramic video diffusion model that employs scene mesh renders as condition, to enable high-quality and geometrically consistent scene video generation. To lift the panorama scene video to 3D world, we propose two separate methods: (1) a feed-forward large panorama reconstruction model for rapid 3D scene reconstruction and (2) an optimization-based pipeline for accurate and detailed 3D scene reconstruction. To facilitate effective training, we also introduce the Matrix-Pano dataset, the first large-scale synthetic collection comprising 116K high-quality static panoramic video sequences with depth and trajectory annotations. Extensive experiments demonstrate that our proposed framework achieves state-of-the-art performance in panoramic video generation and 3D world generation. See more in https://matrix-3d.github.io.