Fast_Human_Pose_Estimation_Pytorch

Fast_Human_Pose_Estimation_Pytorch

权重：model_best.pth.tar

python tools/mpii_demo.py -a hg -s 2 -b 1 --mobile True --checkpoint model_best.pth.tar --in_res 256 --device cuda

效果不是特别好：s 8 b 1 mobile false

  

   

    

     

    
    

     
 import argparse
     
    
   

    

     

    
    

     

       parser = argparse.ArgumentParser(description='PyTorch ImageNet Training')
     
    
   

    

     

    
    

     
 # Model structure
     
    
   

    

     

    
    

     

       parser.add_argument('--arch', '-a', metavar='ARCH', default='hg',
     
    
   

    

     

    
    

     

       choices=model_names,
     
    
   

    

     

    
    

     

       help='model architecture: ' +' | '.join(model_names) +' (default: resnet18)')
     
    
   

    

     

    
    

     

       parser.add_argument('-s', '--stacks', default=8, type=int, metavar='N',
     
    
   

    

     

    
    

     

       help='Number of hourg
     
    
  
 

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/100574661