StereoWorld: Geometry-Aware Monocular-to-Stereo Video Generation

TL;DR

StereoWorld is a novel framework that converts monocular videos into high-quality stereo videos by leveraging geometry-aware regularization and a large HD stereo dataset, significantly improving visual fidelity and 3D structural accuracy.

Contribution

It introduces an end-to-end monocular-to-stereo video generation method with geometry-aware supervision and a new high-definition stereo dataset for training and evaluation.

Findings

Outperforms prior methods in visual fidelity

Ensures geometric consistency in generated stereo videos

Supports high-resolution synthesis with efficient tiling

Abstract

The growing adoption of XR devices has fueled strong demand for high-quality stereo video, yet its production remains costly and artifact-prone. To address this challenge, we present StereoWorld, an end-to-end framework that repurposes a pretrained video generator for high-fidelity monocular-to-stereo video generation. Our framework jointly conditions the model on the monocular video input while explicitly supervising the generation with a geometry-aware regularization to ensure 3D structural fidelity. A spatio-temporal tiling scheme is further integrated to enable efficient, high-resolution synthesis. To enable large-scale training and evaluation, we curate a high-definition stereo video dataset containing over 11M frames aligned to natural human interpupillary distance (IPD). Extensive experiments demonstrate that StereoWorld substantially outperforms prior methods, generating stereo videos with superior visual fidelity and geometric consistency. The project webpage is available at https://ke-xing.github.io/StereoWorld/.