retinaface人脸对齐

https://github.com/foamliu/MobileFaceNet-PyTorch

调用:

从左到右，从上到下，依次5个点

这个需要的是facial5points 格式是：

[x1,x2,x3,x4,x5][y1,y2,y3,y4,y5]

  

   

    

     

    
    

     

      def align_face(img_fn, facial5points):
     
    
   

    

     

    
    

     

       raw = cv2.imread(img_fn, True)  # BGR
     
    
   

    

     

    
    

     

       facial5points = np.reshape(facial5points, (2, 5))
     
    
   

    

     

    
    

     
 
     
    
   

    

     

    
    

     

       crop_size = (image_h, image_w)
     
    
   

    

     

    
    

     
 
     
    
   

    

     

    
    

     

       default_square = True
     
    
   

    

     

    
    

     

       inner_padding_factor = 0.25
     
    
   

    

     

    
    

     

       outer_padding = (0, 0)
     
    
   

    

     

    
    

     

       output_size = (image_h, image_w)
     
    
   

    

     

    
    

     
 
     
    
   

    

     

    
    

     
 # get the reference 5 landmarks position in the crop settings
     
    
   

    

     

    
    

     

       reference_5pts = get_reference_facial_points(
     
    
  
 

文章来源: blog.csdn.net，作者：网奇，版权归原作者所有，如需转载，请联系作者。

原文链接：blog.csdn.net/jacke121/article/details/104941564