# FRDiff

**Repository Path**: koalaaaaaaaaa/FRDiff

## Basic Information

- **Project Name**: FRDiff
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2025-01-12
- **Last Updated**: 2025-01-12

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# FRDiff : Feature Reuse for Universal Training-free Acceleration of Diffusion Models [[Arxiv]](https://arxiv.org/abs/2312.03517) [[Project]](https://jungwon-lee.github.io/Project_FRDiff) [[Colab]](https://colab.research.google.com/drive/1nG15sCcIS-XaZKDvGugBvg4eKF3qftoA#scrollTo=zvA_neljsaaU)
### [Junhyuk So*](https://github.com/junhyukso), [Jungwon Lee*](https://github.com/Jungwon-Lee) and Eunhyeok Park

This repository is official code of [FRDiff : Feature Reuse for Universal Training-free Acceleration of Diffusion Models](https://arxiv.org/abs/2312.03517)


## Overview 
In our work, we introduce an advanced acceleration technique that leverages the ___temporal redundancy___ inherent in diffusion models. Reusing feature maps with high temporal similarity opens up a new opportunity to save computation resources without compromising output quality. To realize the practical benefits of this intuition, we conduct an extensive analysis and propose a novel method, ___FRDiff___. ___FRDiff___ is designed to harness the advantages of both reduced NFE and feature reuse, achieving a Pareto frontier that balances fidelity and latency trade-offs in various generative tasks. 

<p align="center">
<img src=assets/main.png />
</p>


## Experiments
<p align="center">
<img src=assets/result.png />
</p>

<br/>

## Stable Diffusion

### Requirements
```
pip install torch transformers accelerate
pip install diffusers==0.26.3
```

### Sampling
To enjoy FRDiff on Stable Diffusion (SDXL), run following script.
```
cd Stable-Diffusion
python sample.py --num_steps 50 --interval 2 --prompt "a photo of an astronaut on a moon"
```

## DiT

### Requirements
```
conda env create --file DiT/environment.yml
```

### AutoFR Training & Sampling
To run AutoFR training, use following script.
```
cd DiT
python sample.py --lr 5e-3 --wgt 1e-3
```

The trained keyframeset and generated sample results will be saved in directory.  

To use Uniform Keyframeset, uncomment line 137-139 in DiT/model.py

<br/>
       
## Citation
```
@article{so2023frdiff,
         title={FRDiff: Feature Reuse for Exquisite Zero-shot Acceleration of Diffusion Models},
         author={So, Junhyuk and Lee, Jungwon and Park, Eunhyeok},
         journal={arXiv preprint arXiv:2312.03517},
         year={2023}
}
```