pytorchtransform数据处理转c++问题

2023-02-02 12:03

短信预约 -IT技能 免费直播动态提醒

pytorch transform数据处理转c++

python推理代码转c++ sdk过程遇到pytorch数据处理的转换

1.python代码

import torch
from PIL import Image
from torchvision import transforms

data_transform = transforms.Compose(
     [transforms.Resize(256),
      transforms.CenterCrop(224),
      transforms.ToTensor(),
      transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])])

 img = Image.open(img_path)
 img = data_transform(img)

2.transforms.Resize(256)

Parameters
size (sequence or int) –
Desired output size. If size is a sequence like (h, w), output size will be matched to this. If size is an int, smaller edge of the image will be matched to this number. i.e, if height > width, then image will be rescaled to (size * height / width, size).

3.transforms.ToTensor()

Convert a PIL Image or numpy.ndarray to tensor. This transform does not support torchscript.
Converts a PIL Image or numpy.ndarray (H x W x C) in the range [0, 255] to a torch.FloatTensor of shape (C x H x W) in the range [0.0, 1.0] if the PIL Image belongs to one of the modes (L, LA, P, I, F, RGB, YCbCr, RGBA, CMYK, 1) or if the numpy.ndarray has dtype = np.uint8

cv::Mat ClsSixPrivate::processImage(cv::Mat &img) {
    int inW = img.cols;
    int inH = img.rows;
    cv::Mat croped_image;
    if (inW > inH)
    {
        int newWidth = 256 * inW / inH;
        cv::resize(img, img, cv::Size(newWidth, 256), 0, 0, cv::INTER_LINEAR);
        croped_image = img(cv::Rect((newWidth - 224) / 2, 16, 224, 224)).clone();
    }
    else {
        int newHeight= 256 * inH / inW;
        cv::resize(img, img, cv::Size(256, newHeight), 0, 0, cv::INTER_LINEAR);
        croped_image = img(cv::Rect(16, (newHeight - 224) / 2, 224, 224)).clone();
    }
    
    std::vector<float> mean_value{ 0.485, 0.456,0.406 };
    std::vector<float> std_value{ 0.229, 0.224, 0.225 }; 
    cv::Mat dst;
    std::vector<cv::Mat> rgbChannels(3);
    cv::split(croped_image, rgbChannels);

    for (auto i = 0; i < rgbChannels.size(); i++)
    {
        rgbChannels[i].convertTo(rgbChannels[i], CV_32FC1, 1.0 / (std_value[i] * 255.0), (0.0 - mean_value[i]) / std_value[i]);
    }

    cv::merge(rgbChannels, dst);
    return dst;
}