PanoWan: Lifting Diffusion Video Generation Models to 360° with Latitude/Longitude-aware Mechanisms
Paper • 2505.22016 • Published • 1
import torch
from diffusers import DiffusionPipeline
from diffusers.utils import export_to_video
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("Wan-AI/Wan2.1-T2V-1.3B", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("YOUSIKI/PanoWan")
prompt = "Stunning panoramic underwater shot of a vibrant coral reef ecosystem brimming with marine life. Colorful fish dart effortlessly among intricate coral formations, soft rays of sunlight filter through the crystal-clear waters, creating mesmerizing patterns on the ocean floor. Wide-angle capturing vivid hues and abundant biodiversity."
output = pipe(prompt=prompt).frames[0]
export_to_video(output, "output.mp4")
PanoWan
Official repository for "PanoWan: Lifting Diffusion Video Generation Models to 360° with Latitude/Longitude-aware Mechanisms"
Text-to-360° Video Generation
Generate panoramic videos from text prompts:
Zero-Shot Applications
Long Video Generation
Generate extended panoramic videos using temporal windowing and seamless blending:
Super Resolution
Enhance low-resolution panoramic videos to 2x resolution:
|
Low Resolution |
High Resolution |
Semantic Editing
Edit panoramic videos with text-guided modifications:
|
Original |
Edited |
Video Outpainting
Transform conventional videos to panoramic format:
Dataset
The metadata for our dataset is released at HuggingFace.
Citation
@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}
}
- Downloads last month
- 35
Model tree for YOUSIKI/PanoWan
Base model
Wan-AI/Wan2.1-T2V-1.3B