DiffCollage

KO | EN

gitlite — search

#hacktoberfest #react #ai #typescript #llm #linux #nodejs #java #awesome #kubernetes #awesome-list #pytorch #security #open-source

(★ 25)

An unoffical but faithful reproducing of DiffCollage

DiffCollage Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# DiffCollage: Parallel Generation of Large Content with Diffusion Models

<div align="center">
 
 <a href="https://qsh-zh.github.io/" target="_blank">
 Qinsheng Zhang
 &nbsp;1
 &nbsp;2
 </a>
 &emsp;
 &middot;
 &emsp;
 <a href="https://tsong.me/" target="_blank">
 Jiaming Song
 &nbsp;2
 </a>
 &emsp;
 &middot;
 &emsp;
 <a href="https://xunhuang.me/" target="_blank">
 Xun Huang
 &nbsp;2
 </a>
 &emsp;
 &middot;
 &emsp;
 <a href="https://yongxin.ae.gatech.edu/" target="_blank">
 Yongxin Chen
 &nbsp;1
 </a>
 &emsp;
 &middot;
 &emsp;
 <a href="http://mingyuliu.net/" target="_blank">
 Ming-Yu Liu
 &nbsp;2
 </a>

<div>
 <a href="http://www.gatech.edu/" target="_blank" style="display: inline; color: #B31B1B;">1. Georgia Institute of Technology</a>
 &emsp;
 &middot;
 &emsp;
 <a href="https://www.nvidia.com/" target="_blank" style="display: inline; margin-left: 10px; color: #76b900;">2. NVIDIA Corporation</a>
 </div>

<a href="https://arxiv.org/abs/2206.05564" target="_blank">Paper</a>
 &emsp;
 &middot;
 &emsp;
 <a href="https://research.nvidia.com/labs/dir/diffcollage/" target="_blank">Project page</a>
</div>

**TLDR**: DiffCollage, a scalable probabilistic model that synthesizes large content in parallel, including long images, looped motions, and 360 images, with diffusion models only trained on pieces of the content.

![teaser](https://raw.githubusercontent.com/sbyebss/DiffCollage/main/assets/teaser.jpg)

# Diff Collage

Diffusion models and notations in this code base follows [EDM](https://arxiv.org/abs/2206.00364).

* $\sigma = t$
* Diffusion forward process follows $x_\sigma = x_0 + \sigma \epsilon$
* Diffusion training objective $|x_0 - x_\theta(x_0 + \sigma \epsilon, \sigma)|^2$ (data prediction model) or $|\epsilon - \epsilon_\theta(x_0 + \sigma \epsilon, \sigma)|^2$ (noise prediction model)
* Conversion between data prediction model and noise prediction model: $\epsilon_\theta(x_\sigma, \sigma) = \frac{x_\sigma - x_\theta(x_0 + \sigma \epsilon, \sigma)}{\sigma} $

## Caveats

Please be aware of the following points when using this software:

1. **Model Conversion**:
   - If your model has been trained using methods other than [EDM](https://github.com/NVlabs/edm), you may need to convert it to EDM using the [change-of-variable](https://github.com/crowsonkb/k-diffusion/blob/master/k_diffusion/external.py#L116) method.

2. **Sampling hyperparameters**
    - We find stochastic sampling algorithms performance way more better than deterministic sampling algorithm when sampling step is high.

## usage

```python

import diff_collage as dc

def test(eps_fn, s_churn=10.0):
    n_step = 40 # sampling step
    overlap_size = 32 # how much overlap
    num_img = 11 # how many square images
    batch_size = 5
    ts_order = 5 # sampling timestamp schedule
    img_shape = (3, 64, 64) # image shape

# sampling with conditional independence assumption
    worker = dc.condind_long.CondIndLong(img_shape, eps_fn, num_img, overlap_size=overlap_size)
    sample = dc.sampling(
        x = worker.generate_xT(batch_size),
        noise_fn = worker.noise,
        rev_ts = worker.rev_ts(n_step, ts_order),
        x0_pred_fn = worker.x0_fn,
        s_churn = s_churn,
        is_traj = False # return sampling traj or not
    )

# sampling with average noise method
    worker = dc.AvgLong(img_shape, eps_fn, num_img, overlap_size=overlap_size)
    sample = dc.sampling(
        x = worker.generate_xT(batch_size),
        noise_fn = worker.noise,
        rev_ts = worker.rev_ts(n_step, ts_order),
        x0_pred_fn = worker.x0_fn,
        s_churn = s_churn,
        is_traj = False # return sampling traj or not
    )
```

# Application : text-to-motion generation

**The repo provides demo code for looped motion generation based on [Human Motion Diffusion Model](https://guytevet.github.io/mdm-page/) pretrained models**

## seutp environment

```shell
sudo apt update
sudo apt install ffmpeg

conda env create -f environment.yml
conda activate mdm
python -m spacy download en_core_web_sm
pip install git+https://github.com/openai/CLIP.git

bash prepare/download_smpl_files.sh
bash prepare/download_glove.sh
```

## download checkpoint

```shell
cd save
gdown "https://drive.google.com/u/0/uc?id=1PE0PK8e5a5j-7-Xhs5YET5U5pGh0c821&export=download&confirm=t"
unzip humanml_trans_enc_512.zip
cd ..

# sanity check
python -m sample --model_path ./save/humanml_trans_enc_512/model000200000.pt --text_prompt "the person walked forward and is picking up his toolbox."
```

# Reference

```tex
@inproceedings{zhange2023diffcollage,
    title={DiffCollage: Parallel Generation of Large Content with Diffusion Models},
    author={Qinsheng Zhang and Jiaming Song and Xun Huang and Yongxin Chen and Ming-yu Liu},
    booktitle={CVPR},
    year={2023}
}
```

DiffCollage

DiffCollage Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

DiffCollage

DiffCollage Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README