Search papers, labs, and topics across Lattice.
This study introduces PanoWorld, a novel approach to long-range memory in panoramic world models that leverages the rotation-equivariant properties of omnidirectional representations. By simplifying camera trajectories and employing Dense Panoramic Ray-Conditioning (DPRC) alongside Geometry-aware Memory Augmentation (GMA), the authors create a robust framework for modeling current actions and enhancing memory retention. Extensive evaluations on the newly constructed World360 dataset reveal that PanoWorld significantly outperforms existing methods in terms of physical consistency across varying spatial and illumination conditions.
PanoWorld achieves unprecedented performance in panoramic generation by effectively integrating long-range memory with rotation-equivariant representations.
In this work, we aim to address the challenge of long-range memory in panoramic world models by exploiting the rotation-equivariant property of omnidirectional representations, where rotation can be treated as an implicit geometric transformation.Building on this insight, we propose PanoWorld, which simplifies camera trajectories into translations via fixed headings for both current-action modeling and long-range memory through Dense Panoramic Ray-Conditioning (DPRC) and Geometry-aware Memory Augmentation (GMA).Then, a three-stage training pipeline is introduced to progressively optimize each component. To better evaluate physical consistency under large-scale spatial variations and diverse illumination conditions, where existing datasets are relatively stable, we construct World360, a large-scale dataset consisting of both real-world video clips collected via panoramic unmanned aerial vehicles and high-quality simulated clips generated by AirSim360.Extensive experiments on World360 demonstrate the effectiveness of PanoWorld, outperforming alternative methods by a large margin.Our models, training code, and dataset will be publicly available. More information can be found on our project page: https://lihaoy-ux.github.io/panoworld-page/.