Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
Manuel Garcia 84c8958dae | 2 years ago | |
---|---|---|
.. | ||
blazeface | 2 years ago | |
demo_images | 3 years ago | |
element_source | 3 years ago | |
solov2 | 2 years ago | |
solov2_blazeface | 2 years ago | |
README.md | 3 years ago | |
test_main.py | 2 years ago | |
test_server.py | 3 years ago |
通过SOLOv2实例分割模型分割人像,并通过BlazeFace关键点模型检测人脸关键点,然后根据两个模型输出结果更换圣诞风格背景并为人脸加上圣诞老人胡子、圣诞眼镜及圣诞帽等特效。本项目通过PaddleHub可直接发布Server服务,供本地调试与前端直接调用接口。您可通过以下二维码中微信小程序直接体验:
paddlepaddle >= 2.0.0rc0
paddlehub >= 2.0.0b1
首先要获取模型,可在模型配置文件里配置solov2
与blazeface_keypoint
,训练模型,并导出模型。也可直接下载我们准备好模型:
blazeface_keypoint模型和
solov2模型。
注意: 下载的模型需要解压后使用。
然后将两个模型文件夹中的文件(infer_cfg.yml
、__model__
和__params__
)分别拷贝至blazeface/blazeface_keypoint/
和 solov2/solov2_r101_vd_fpn_3x/
文件夹内。
hub install solov2
hub install blazeface
$ hub install solov2_blazeface
python test_main.py
运行成功后,预测结果会保存到chrismas_final.png
。
export CUDA_VISIBLE_DEVICES=0
hub serving start -m solov2_blazeface -p 8880
python test_server.py
运行成功后,预测结果会保存到chrismas_final.png
。
PaddleDetection
Python Jupyter Notebook Markdown C++ Java other
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》