Depth-Anything-V2_Safetensors
Introduction
Depth Anything V2 is trained from 595K synthetic labeled images and 62M+ real unlabeled images, providing the most capable monocular depth estimation (MDE) model with the following features:
- more fine-grained details than Depth Anything V1
- more robust than Depth Anything V1 and SD-based models (e.g., Marigold, Geowizard)
- more efficient (10x faster) and more lightweight than SD-based models
- impressive fine-tuned performance with our pre-trained models
- models have been converted into .safetensors
Installation
git clone https://github.com/MackinationsAi/Upgraded-Depth-Anything-V2.git
cd Upgraded-Depth-Anything-V2
one_click_install.bat
Usage
- Download the Depth-Anything-V2-Large model | 654.9M | Download | first and put it under the
checkpoints
directory. - Download the Depth-Anything-V2-Base model | 190.4M | Download | second and put it under the
checkpoints
directory. - Download the Depth-Anything-V2-Small model | 48.4M | Download | third and put it under the
checkpoints
directory. - Download the Depth-Anything-V2-Giant model | 1.3B | Coming soon (Still not available) | Download Doesn't Work - Model is still a WIP | fourth and put it under the
checkpoints
directory.
Citation
If you find this project useful, please consider citing below, give these converted models & upgraded linked repo a star/follow & share it w/ others in the community!
@article{depth_anything_v2,
title={Depth Anything V2},
author={Yang, Lihe and Kang, Bingyi and Huang, Zilong and Zhao, Zhen and Xu, Xiaogang and Feng, Jiashi and Zhao, Hengshuang},
journal={arXiv:2406.09414},
year={2024}
}
@inproceedings{depth_anything_v1,
title={Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data},
author={Yang, Lihe and Kang, Bingyi and Huang, Zilong and Xu, Xiaogang and Feng, Jiashi and Zhao, Hengshuang},
booktitle={CVPR},
year={2024}
}