metadata
base_model: stabilityai/stable-diffusion-2
base_model_relation: finetune
language:
- en
library_name: diffusers
license: creativeml-openrail-m
tags:
- stereo generation
- augmented reality
- image-to-image
- diffusion
- zero-shot
StereoSpace Model Card
This is a model card for the stereospace-v1-0 model for baseline-controllable single-image stereo view generation.
The model is derived from stabilityai/stable-diffusion-2 using fine-tuning as described in our paper titled
"StereoSpace: Depth-Free Synthesis of Stereo Geometry via End-to-End Diffusion in a Canonical Space" by
Tjark Behrens, Anton Obukhov, Bingxin Ke, Fabio Tosi, Matteo Poggi, Konrad Schindler.
See the Quick Start section of the paper's code repository for instructions on how to set up the environment and process images with this model.
- Model Name:
stereospace-v1-0 - Task: Single Image Stereo Image Generation
- Base Model:
stabilityai/stable-diffusion-2 - Model Type: End-to-end latent diffusion baseline-controllable single-image stereo view generation.
- Resources for more information: Project Website, Paper, Code.
- Framework: PyTorch, Diffusers
- Language: English.
- License: CreativeML Open RAIL++-M License.
- Developed by: PRS-ETH
- Cite as:
@misc{behrens2025stereospace,
title = {StereoSpace: Depth-Free Synthesis of Stereo Geometry via End-to-End Diffusion in a Canonical Space},
author = {Tjark Behrens and Anton Obukhov and Bingxin Ke and Fabio Tosi and Matteo Poggi and Konrad Schindler},
year = {2025},
eprint = {2512.10959},
archivePrefix= {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/2512.10959},
}