Generative Photography: Scene-Consistent Camera Control for Realistic Text-to-Image Synthesis
Paper • 2412.02168 • Published • 1
import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("pandaphd/generative_photography", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]Generative Photography
Project Page | Paper | Github
-----
Generative Photography: Scene-Consistent Camera Control for Realistic Text-to-Image Synthesis
In this repository, we present Generative Photography, a new
🔥 Latest News!!
- March 3, 2025: Release offical code and pre-trained weights.
- Feb 26, 2025: Paper is accepted by CVPR 2025!
- Dec 20, 2024: Release dataset.
Citation
If you find our work helpful, please cite us.
@article{Yuan_2024_GenPhoto,
title={Generative Photography: Scene-Consistent Camera Control for Realistic Text-to-Image Synthesis},
author={Yuan, Yu and Wang, Xijun and Sheng, Yichen and Chennuri, Prateek and Zhang, Xingguang and Chan, Stanley},
journal={CVPR},
year={2025}
}
- Downloads last month
- -