使用OpenCV进行图片风格化

ゞ浴缸里的玫瑰 2023-01-22 15:00 5阅读 0赞

OpenCV现在可以直接读取训练好的风格化的深度学习模型，模型下载地址如下：

[https://github.com/jcjohnson/fast-neural-style/blob/master/models/download\_style\_transfer\_models.sh][https_github.com_jcjohnson_fast-neural-style_blob_master_models_download_style_transfer_models.sh]

话不多说，直接先上代码：

这里我是使用的其中一个模型

import cv2 as cv
    import numpy as np
    
    
    net = cv.dnn.readNetFromTorch('feathers.t7')
    print(net)
    frame=cv.imread('5656.jpg')
    inWidth, inHeight=frame.shape[1],frame.shape[0]
    
    inp = cv.dnn.blobFromImage(frame, 1.0, (inWidth, inHeight),
                                  (103.939, 116.779, 123.68), swapRB=False, crop=False)
    
    net.setInput(inp)
    out = net.forward()
    print(out)
    print(out.shape)
    out = out.reshape(3, out.shape[2], out.shape[3])
    out[0] += 103.939
    out[1] += 116.779
    out[2] += 123.68
    
    out = out.transpose(1, 2, 0)
    print(out)
    cv.imwrite('out.jpg',out)

看一下运行的效果：

原图：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3pob3VfNDM4_size_16_color_FFFFFF_t_70][]

风格化后的结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3pob3VfNDM4_size_16_color_FFFFFF_t_70 1][]

推理速度也很快，没有使用gpu也很快

刚才是试的feathers，现试一下candy这个模型：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3pob3VfNDM4_size_16_color_FFFFFF_t_70 2][]

还是挺漂亮的

[https_github.com_jcjohnson_fast-neural-style_blob_master_models_download_style_transfer_models.sh]: https://github.com/jcjohnson/fast-neural-style/blob/master/models/download_style_transfer_models.sh
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3pob3VfNDM4_size_16_color_FFFFFF_t_70]: /images/20221020/afadc3310c904a4fac1c72f31ada8b9e.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3pob3VfNDM4_size_16_color_FFFFFF_t_70 1]: /images/20221020/1bd61cf9b4e4495bac6c92b392460090.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3pob3VfNDM4_size_16_color_FFFFFF_t_70 2]: /images/20221020/4a3c0d7c34364fc59364c95c3af724f7.png