---
library_name: diffusers
license: apache-2.0
---

# Lotus: Diffusion-based Visual Foundation Model for High-quality Dense Prediction

<!-- Provide a quick summary of what the model is/does. -->
This model belongs to the family of official Lotus models. </br>
Some training normals in the Hypersim dataset are not properly oriented towards the camera. This models was re-trained using aligned surface normals, referred to [GeoWizard](https://github.com/fuxiao0719/GeoWizard/blob/5ff496579c6be35d9d86fe4d0760a6b5e6ba25c5/geowizard/training/dataloader/file_io.py#L79), and achieves significantly improved results. 

[![Paper](https://img.shields.io/badge/Project-Website-pink?logo=googlechrome&logoColor=white)](https://lotus3d.github.io/)
[![Paper](https://img.shields.io/badge/arXiv-Paper-b31b1b?logo=arxiv&logoColor=white)](https://arxiv.org/abs/2409.18124)
[![HuggingFace Demo](https://img.shields.io/badge/🤗%20HuggingFace-Demo-yellow)](https://huggingface.co/spaces/haodongli/Lotus)
[![GitHub](https://img.shields.io/github/stars/EnVision-Research/Lotus?style=default&label=GitHub%20★&logo=github)](https://github.com/EnVision-Research/Lotus)

Developed by: 
[Jing He](https://scholar.google.com/citations?hl=en&user=RsLS11MAAAAJ)<span style="color:red;">&#10033;</span>,
[Haodong Li](https://haodong-li.com/)<span style="color:red;">&#10033;</span>,
[Wei Yin](https://yvanyin.net/),
[Yixun Liang](https://yixunliang.github.io/),
[Leheng Li](https://len-li.github.io/),
[Kaiqiang Zhou](),
[Hongbo Zhang](),
[Bingbing Liu](https://scholar.google.com/citations?user=-rCulKwAAAAJ&hl=en),
[Ying-Cong Chen](https://www.yingcong.me/)&#9993;

![teaser](assets/badges/teaser_1.jpg)
![teaser](assets/badges/teaser_2.jpg)

## Usage
Please refer to this [page](https://github.com/EnVision-Research/Lotus).