PanoWan: Lifting Diffusion Video Generation Models to 360° with Latitude/Longitude-aware Mechanisms
Paper
•
2505.22016
•
Published
Official repository for "PanoWan: Lifting Diffusion Video Generation Models to 360° with Latitude/Longitude-aware Mechanisms"
Generate panoramic videos from text prompts:
Generate extended panoramic videos using temporal windowing and seamless blending:
Enhance low-resolution panoramic videos to 2x resolution:
|
Low Resolution |
High Resolution |
Edit panoramic videos with text-guided modifications:
|
Original |
Edited |
Transform conventional videos to panoramic format:
The metadata for our dataset is released at HuggingFace.
@article{xia2025panowan,
title={PanoWan: Lifting Diffusion Video Generation Models to 360° with Latitude/Longitude-aware Mechanisms},
author={Xia, Yifei and Weng, Shuchen and Yang, Siqi and Liu, Jingqi and Zhu, Chengxuan and Teng, Minggui and Jia, Zijian and Jiang, Han and Shi, Boxin},
journal={arXiv preprint arXiv:2505.22016},
year={2025}
}
Base model
Wan-AI/Wan2.1-T2V-1.3B