wuyefeilin

1. In 1024 resolution， ppmatting is better than modnet. 3. Maybe you need a dataset firstly. 4. Maybe you can train the model by combine the studio quality images and...

> Thanks for the reply, I do have a dataset of 80k anotated images.How do i fix the multiple class object matting? There's too much data to know the distribution...

Sorry, but I Do not receive your mail. If you don't mind, please resent it.

目前不支持，您可以按照蒸馏的形式自行进行改造哈

May be you can try the `PPLiteSeg`

Yes, we have plan to support the C++ inference in the future. The [url](https://github.com/PaddlePaddle/PaddleSeg/tree/457af2edb51a3c85ac9fd019e564c4099b1ee8bd/deploy/onnxruntime_cpp) can not be used directly. You can make changes as needed。

You can refer the python deployment of matting to adjust the cpp_inference.

看提示是因为输入尺寸不匹配的导致的，确定没有修改过代码吗

> They have shared only the inference weights, which is a zip which contains deploy.yaml as config which you can rub by using the file deploy/python/infer.py Yes， he is right....

Coming soon.