English

PanFlow: Decoupled Motion Control for Panoramic Video Generation

Computer Vision and Pattern Recognition 2025-12-02 v1

Abstract

Panoramic video generation has attracted growing attention due to its applications in virtual reality and immersive media. However, existing methods lack explicit motion control and struggle to generate scenes with large and complex motions. We propose PanFlow, a novel approach that exploits the spherical nature of panoramas to decouple the highly dynamic camera rotation from the input optical flow condition, enabling more precise control over large and dynamic motions. We further introduce a spherical noise warping strategy to promote loop consistency in motion across panorama boundaries. To support effective training, we curate a large-scale, motion-rich panoramic video dataset with frame-level pose and flow annotations. We also showcase the effectiveness of our method in various applications, including motion transfer and video editing. Extensive experiments demonstrate that PanFlow significantly outperforms prior methods in motion fidelity, visual quality, and temporal coherence. Our code, dataset, and models are available at https://github.com/chengzhag/PanFlow.

Keywords

Cite

@article{arxiv.2512.00832,
  title  = {PanFlow: Decoupled Motion Control for Panoramic Video Generation},
  author = {Cheng Zhang and Hanwen Liang and Donny Y. Chen and Qianyi Wu and Konstantinos N. Plataniotis and Camilo Cruz Gambardella and Jianfei Cai},
  journal= {arXiv preprint arXiv:2512.00832},
  year   = {2025}
}

Comments

Accepted by AAAI. Code: https://github.com/chengzhag/PanFlow

R2 v1 2026-07-01T08:01:41.137Z