# MMPose for AIGC (AI Generated Content)
English | [简体中文](./README_CN.md)
This project will demonstrate how to use MMPose to generate skeleton images for pose guided AI image generation.
Currently, we support:
- [T2I Adapter](https://huggingface.co/spaces/Adapter/T2I-Adapter)
Please feel free to share interesting pose-guided AIGC projects to us!
## Get Started
### Generate OpenPose-style Skeleton
#### Step 1: Preparation
Run the following commands to prepare the project:
```shell
# install mmpose mmdet
pip install openmim
git clone https://github.com/open-mmlab/mmpose.git
cd mmpose
mim install -e .
mim install "mmdet>=3.0.0rc6"
# download models
bash download_models.sh
```
#### Step 2: Generate a Skeleton Image
Run the following command to generate a skeleton image:
```shell
# generate a skeleton image
bash mmpose_openpose.sh ../../tests/data/coco/000000000785.jpg
```
The input image and its skeleton are as follows:
### Generate MMPose-style Skeleton
#### Step 1: Preparation
**Env Requirements:**
- GCC >= 7.5
- cmake >= 3.14
Run the following commands to install the project:
```shell
bash install_posetracker_linux.sh
```
After installation, files are organized as follows:
```shell
|----mmdeploy-1.0.0-linux-x86_64-cxx11abi
| |----README.md
| |----rtmpose-ort
| | |----rtmdet-nano
| | |----rtmpose-m
| | |----000000147979.jpg
| | |----t2i-adapter_skeleton.txt
```
#### Step 2: Generate a Skeleton Image
Run the following command to generate a skeleton image:
```shell
# generate a skeleton image
bash mmpose_style_skeleton.sh \
mmdeploy-1.0.0-linux-x86_64-cxx11abi/rtmpose-ort/000000147979.jpg
```
For more details, you can refer to [RTMPose](../rtmpose/README.md).
The input image and its skeleton are as follows:
### Upload to T2I-Adapter
The demo page of T2I- Adapter is [Here](https://huggingface.co/spaces/Adapter/T2I-Adapter).
[](https://huggingface.co/spaces/ChongMou/T2I-Adapter)
## Gallery
