🏆 [ICCV 2025] ZeroStereo: Zero-shot Stereo Matching from Single Images 🏆

Xianqi Wang, Hao Yang, Gangwei Xu, Junda Cheng, Min Lin, Yong Deng, Jinliang Zang, Yurui Chen, Xin Yang

Huazhong University of Science and Technology, Autel Robotics, Optics Valley Laboratory

Used title: StereoGen: High-quality Stereo Image Generation from a Single Image

🔄 Update

• 09/16/2025: Update MfS35K to Hugging Face.
• 09/16/2025: Update MfS35K to Baidu Netdisk.
• 07/29/2025: Update the fine-tuning code for SDv2I.
• 07/14/2025: Update the generation code to improve the quality of the right image edges.

⚙️ Environment

conda create -n zerostereo python=3.12
conda activate zerostereo

pip install tqdm numpy wandb opt_einsum hydra-core
pip install scipy torch torchvision diffusers transformers opencv-python matplotlib
pip install xformers accelerate scikit-image timm==0.5.4

📂 Required Data

Data for fine-tuning:

• TartanAir

• CREStereo Dataset

• Falling Things

• Scene Flow

• VKITTI 2

Data for generation:

• DiW

• COCOStuff

• DIODE

• Mapillary

• ADE20K

The filepath format should be consistent with the filelist.

Data for training:

• MfS35K on Baidu Netdisk

• MfS35K on Hugging Face

Data for evaluation:

• KITTI 2015

• Middlebury

• ETH3D

🎁 Pre-Trained Model

Model	Link
SDv2I	Download 🤗
StereoGen	Download 🤗
Zero-RAFT-Stereo	Download 🤗
Zero-IGEV-Stereo	Download 🤗

The link to the original SDv2I is invalid. Please use the copy from others, like Download 🤗.

🚀 Fine-tuning

To fine-tune SDv2I, run:

accelerate launch train_stereogen.py

🛠️ Generation

To generate MfS35K, run:

accelerate launch generate_mono.py
accelerate launch generate_stereo.py

🚀 Training

To train Zero-RAFT-Stereo and Zero-IGEV-Stereo, run:

CUDA_VISIBLE_DEVICES='0,1' accelerate launch train_stereo.py

📊 Evaluation

To evaluate Zero-RAFT-Stereo, run:

accelerate launch evaluate_stereo.py

To evaluate Zero-IGEV-Stereo, modify config/evaluate_stereo.yaml or run:

accelerate launch evaluate_stereo.py model=igev_stereo checkpoint=checkpoint/igev_stereo/model_700.safetensors  

🎥 Demo

To save disparity, run:

accelerate launch save_disparity.py

🔔 Notification

🙏 Acknowledgement

This project is based on MfS-Stereo, Depth Anything V2, Marigold, RAFT-Stereo, and IGEV-Stereo. We thank the original authors for their excellent works.