---
license: mit
---

# Doduo: Dense Visual Correspondence from Unsupervised Semantic-Aware Flow

[Zhenyu Jiang](http://zhenyujiang.me), [Hanwen Jiang](https://hwjiang1510.github.io/), [Yuke Zhu](https://www.cs.utexas.edu/~yukez/)

University of Texas at Austin

## Usage

```python
from transformers import AutoModel
model = AutoModel.from_pretrained("stevetod/doduo", trust_remote_code=True)
flow = model(frame_src, frame_dst)
```