Pytorch实现人脸多属性识别

数据来源：CalebA人脸数据集（官网链接）是香港中文大学的开放数据，包含10,177个名人身份的202,599张人脸图片，并且都做好了特征标记，这对人脸相关的训练是非常好用的数据集。共计40个特征，具体是哪些特征，可以去官网查询。话不多说，直接开始流程。

整个流程可以分为大致以下几个步骤：

1.图片预处理

2.构建网络

3.训练

4.测试

5.优化

一。图片加载，以为源数据没有经过处理，我们要重写torch.utils.data.Dataloader()处理图片，然后才能将图片用于加载。代码如下：

def default_loader(path):try:img = Image.open(path)return img.convert('RGB')except:print("Can not open {0}".format(path))
class myDataset(Data.DataLoader):def __init__(self,img_dir,img_txt=img_txt,transform=None,loader=default_loader):img_list = []img_labels = []fp = open(img_txt,'r')for line in fp.readlines():if len(line.split())!=41:continueimg_list.append(line.split()[0])img_label_single = []for value in line.split()[1:]:if value == '-1':img_label_single.append(0)if value == '1':img_label_single.append(1)img_labels.append(img_label_single)self.imgs = [os.path.join(img_dir,file) for file in img_list]self.labels = img_labelsself.transform = transform self.loader = loader def __len__(self):return len(self.imgs)def __getitem__(self,index):img_path = self.imgs[index]label = torch.from_numpy(np.array(self.labels[index],dtype=np.int64))img = self.loader(img_path)if self.transform is not None:try:img = self.transform(img)except:print('Cannot transform image: {}'.format(img_path))return img,label

图片增强、归一化处理和加载：

transform = transforms.Compose([transforms.Resize(40),transforms.CenterCrop(32),transforms.RandomHorizontalFlip(),transforms.ToTensor(),transforms.Normalize(mean=[0.5,0.5,0.5],std = [0.5,0.5,0.5])])

#训练集

train_dataset = myDataset(img_dir=img_root,img_txt=train_txt,transform= transform)train_dataloader = Data.DataLoader(train_dataset,batch_size = batch_size,shuffle=True)

#测试集

test_dataset = myDataset(img_dir=img_root,img_txt = test_txt,transform= transform)
test_dataloader = Data.DataLoader(test_dataset,batch_size = batch_size,shuffle=True)

构建网络：我使用的网络结构是每种属性使用3层卷积加上3层fc。网络结构比较简单，导致准确率不会有太高的表现，如果有兴趣可以做下优化，文末有优化的思路供大家讨论。好了，先上代码：

def make_conv():return nn.Sequential(nn.Conv2d(3,16,3,1,1),nn.ReLU(),nn.MaxPool2d(2),nn.Conv2d(16,32,3,1,1),nn.ReLU(),nn.MaxPool2d(2),nn.Conv2d(32,64,3,1,1),nn.ReLU(),#nn.Dropout(0.5),nn.MaxPool2d(2))
def make_fc():return nn.Sequential(nn.Linear(64*4*4,128),nn.ReLU(),#nn.Dropout(0.5),nn.Linear(128,64),nn.ReLU(),nn.Dropout(0.5),#Dropout()可以一定程度上防止过拟合，放在不同位置或许会有意想不到的结果，有条件可以多尝试几次nn.Linear(64,2))
class face_attr(nn.Module):def __init__(self):super(face_attr,self).__init__()#attr0self.attr0_layer1 = make_conv()self.attr0_layer2 = make_fc()#attr1self.attr1_layer1 = make_conv()self.attr1_layer2 = make_fc()...#每一中属性的计算都是相同的，在文中省略，#attr38self.attr38_layer1 = make_conv()self.attr38_layer2 = make_fc()#attr39self.attr39_layer1 = make_conv()self.attr39_layer2 = make_fc()def forward(self,x):out_list = []#out0out0 = self.attr0_layer1(x)out0 = out0.view(out0.size(0),-1)out0 = self.attr0_layer2(out0)out_list.append(out0)...#out39out39 = self.attr39_layer1(x)out39 = out39.view(out39.size(0),-1)out39 = self.attr39_layer2(out39)out_list.append(out39)return out_list

接下来就可以开始训练网络了，定义优化器的时候可以设置一下weight_decay=1e-8，也可以在一定程度上防止过拟合。

module = face_attr()
#print(module)optimizer = optim.Adam(module.parameters(),lr = 0.001,weight_decay=1e-8)loss_list = []
for i in range(40):loss_func = nn.CrossEntropyLoss()loss_list.append(loss_func)
#loss_func = nn.CrossEntropyLoss()
for Epoch in range(50):all_correct_num = 0for ii,(img,label) in enumerate(train_dataloader):img = Variable(img)label = Variable(label)output = module(img)optimizer.zero_grad()for i in range(40):loss = loss_list[i](output[i],label[:,i])loss.backward()_,predict = torch.max(output[i],1)correct_num = sum(predict==label[:,i])all_correct_num += correct_num.data[0]optimizer.step()        Accuracy =  all_correct_num *1.0/(len(train_dataset)*40.0)print('Epoch ={0},all_correct_num={1},Accuracy={2}'.format(Epoch,all_correct_num,Accuracy))torch.save(module,'W:/pic_data/face/CelebA/Img/face_attr40dro1.pkl')#每跑一个epoch就保存一次模型

测试网络：和训练类似，只是不用优化和做反向传播。

module = torch.load('W:/pic_data/face/CelebA/Img/face_attr40dro1.pkl')#加载刚刚保存的网络
module.eval()#改成测试模式
all_correct_num = 0
for ii,(img,label) in enumerate(test_dataloader):img = Variable(img)label = Variable(label)output = module(img)for i in range(40):_,predict = torch.max(output[i],1)correct_num = sum(predict==label[:,i])all_correct_num += correct_num.data[0]
Accuracy =  all_correct_num *1.0/(len(test_dataset)*40.0)
print('all_correct_num={0},Accuracy={1}'.format(all_correct_num,Accuracy))

总结：我因为是笔记本电脑，没有GPU，所以只给了大约5000个数据用于训练（即使是5000个数据我的电脑也跑了2天才跑完50个epoch），1000个数据用于测试。测试的准确率在90%左右。有条件的同学可以做些优化，下面提供一些可以优化的方面：

1.图片增强，我因为电脑不给力，无法处理较大的数据，所以将原始图片缩放到40*40，然后截取了32*32作为输入，如果有GPU等条件，可以考虑128*128输入

2.最开始没加Dropout()的时候出现了过拟合的情况，当然这也和训练集较小有关系，建议训练集给到60000个样本。上述提到的Dropout()多尝试几个位置，我是放在了最后一层输出之前。期待大家尝试之后分享一下结果。

3.我现在每种属性都是使用相同的网络，但是这种网络有可能不是对于每种属性都是最优的选择，可以针对每一种属性单独写一层网络。例如：某个属性使用全fc层就可以达到很高的准确率，某个属性或许需要4层卷积+2层fc可以达到很好的效果，这种情况只有靠多尝试，算法什么的可能并不能给出哪种才是最适合的网络模型。建议输出每种属性的准确率，然后针对准确率较低的属性做相应的网络优化。

4.多准备几块GPU做训练和测试吧，没GPU真不给力。希望有小伙伴在这个网络的基础上能达到更高的准确率。

附上整体代码如下：

# -*- coding: utf-8 -*-
"""
Created on Sun Jun 17 11:54:36 2018@author: sky-hole
"""import torch
import torch.nn as nn
from torch.autograd import Variable
import torch.optim as optim
import torchvision.transforms as transforms
import torch.utils.data as Data
from PIL import Image
import numpy as np
import osimg_root = 'W:/pic_data/face/CelebA/Img/img_align_celeba'
train_txt = 'W:/pic_data/face/CelebA/Img/train10000.txt'
batch_size = 2def default_loader(path):try:img = Image.open(path)return img.convert('RGB')except:print("Can not open {0}".format(path))
class myDataset(Data.DataLoader):def __init__(self,img_dir,img_txt,transform=None,loader=default_loader):img_list = []img_labels = []fp = open(img_txt,'r')for line in fp.readlines():if len(line.split())!=41:continueimg_list.append(line.split()[0])img_label_single = []for value in line.split()[1:]:if value == '-1':img_label_single.append(0)if value == '1':img_label_single.append(1)img_labels.append(img_label_single)self.imgs = [os.path.join(img_dir,file) for file in img_list]self.labels = img_labelsself.transform = transform self.loader = loader def __len__(self):return len(self.imgs)def __getitem__(self,index):img_path = self.imgs[index]label = torch.from_numpy(np.array(self.labels[index],dtype=np.int64))img = self.loader(img_path)if self.transform is not None:try:img = self.transform(img)except:print('Cannot transform image: {}'.format(img_path))return img,label
transform = transforms.Compose([transforms.Resize(40),transforms.CenterCrop(32),transforms.RandomHorizontalFlip(),transforms.ToTensor(),transforms.Normalize(mean=[0.5,0.5,0.5],std = [0.5,0.5,0.5])])train_dataset = myDataset(img_dir=img_root,img_txt=train_txt,transform= transform)
train_dataloader = Data.DataLoader(train_dataset,batch_size = batch_size,shuffle=True)#print(len(train_dataset))
#print(len(train_dataloader))
def make_conv():return nn.Sequential(nn.Conv2d(3,16,3,1,1),nn.ReLU(),nn.MaxPool2d(2),nn.Conv2d(16,32,3,1,1),nn.ReLU(),nn.MaxPool2d(2),nn.Conv2d(32,64,3,1,1),nn.ReLU(),#nn.Dropout(0.5),nn.MaxPool2d(2))
def make_fc():return nn.Sequential(nn.Linear(64*4*4,128),nn.ReLU(),#nn.Dropout(0.5),nn.Linear(128,64),nn.ReLU(),nn.Dropout(0.5),nn.Linear(64,2))
class face_attr(nn.Module):def __init__(self):super(face_attr,self).__init__()#attr0self.attr0_layer1 = make_conv()self.attr0_layer2 = make_fc()#attr1self.attr1_layer1 = make_conv()self.attr1_layer2 = make_fc()#attr2self.attr2_layer1 = make_conv()self.attr2_layer2 = make_fc()#attr3self.attr3_layer1 = make_conv()self.attr3_layer2 = make_fc()#attr4self.attr4_layer1 = make_conv()self.attr4_layer2 = make_fc()#attr5self.attr5_layer1 = make_conv()self.attr5_layer2 = make_fc()#attr6self.attr6_layer1 = make_conv()self.attr6_layer2 = make_fc()#attr7self.attr7_layer1 = make_conv()self.attr7_layer2 = make_fc()#attr8self.attr8_layer1 = make_conv()self.attr8_layer2 = make_fc()#attr9self.attr9_layer1 = make_conv()self.attr9_layer2 = make_fc()#attr10self.attr10_layer1 = make_conv()self.attr10_layer2 = make_fc()#attr11self.attr11_layer1 = make_conv()self.attr11_layer2 = make_fc()#attr12self.attr12_layer1 = make_conv()self.attr12_layer2 = make_fc()#attr13self.attr13_layer1 = make_conv()self.attr13_layer2 = make_fc()#attr14self.attr14_layer1 = make_conv()self.attr14_layer2 = make_fc()#attr15self.attr15_layer1 = make_conv()self.attr15_layer2 = make_fc()#attr16self.attr16_layer1 = make_conv()self.attr16_layer2 = make_fc()#attr17self.attr17_layer1 = make_conv()self.attr17_layer2 = make_fc()#attr18self.attr18_layer1 = make_conv()self.attr18_layer2 = make_fc()#attr19self.attr19_layer1 = make_conv()self.attr19_layer2 = make_fc()#attr20self.attr20_layer1 = make_conv()self.attr20_layer2 = make_fc()#attr21self.attr21_layer1 = make_conv()self.attr21_layer2 = make_fc()#attr22self.attr22_layer1 = make_conv()self.attr22_layer2 = make_fc()#attr23self.attr23_layer1 = make_conv()self.attr23_layer2 = make_fc()#attr24self.attr24_layer1 = make_conv()self.attr24_layer2 = make_fc()#attr25self.attr25_layer1 = make_conv()self.attr25_layer2 = make_fc()#attr26self.attr26_layer1 = make_conv()self.attr26_layer2 = make_fc()#attr27self.attr27_layer1 = make_conv()self.attr27_layer2 = make_fc()#attr28self.attr28_layer1 = make_conv()self.attr28_layer2 = make_fc()#attr29self.attr29_layer1 = make_conv()self.attr29_layer2 = make_fc()#attr30self.attr30_layer1 = make_conv()self.attr30_layer2 = make_fc()#attr31self.attr31_layer1 = make_conv()self.attr31_layer2 = make_fc()#attr32self.attr32_layer1 = make_conv()self.attr32_layer2 = make_fc()#attr33self.attr33_layer1 = make_conv()self.attr33_layer2 = make_fc()#attr34self.attr34_layer1 = make_conv()self.attr34_layer2 = make_fc()#attr35self.attr35_layer1 = make_conv()self.attr35_layer2 = make_fc()#attr36self.attr36_layer1 = make_conv()self.attr36_layer2 = make_fc()#attr37self.attr37_layer1 = make_conv()self.attr37_layer2 = make_fc()#attr38self.attr38_layer1 = make_conv()self.attr38_layer2 = make_fc()#attr39self.attr39_layer1 = make_conv()self.attr39_layer2 = make_fc()def forward(self,x):out_list = []#out0out0 = self.attr0_layer1(x)out0 = out0.view(out0.size(0),-1)out0 = self.attr0_layer2(out0)out_list.append(out0)#out1out1 = self.attr1_layer1(x)out1 = out1.view(out1.size(0),-1)out1 = self.attr1_layer2(out1)out_list.append(out1)#out2out2 = self.attr2_layer1(x)out2 = out2.view(out2.size(0),-1)out2 = self.attr2_layer2(out2)out_list.append(out2)#out3out3 = self.attr3_layer1(x)out3 = out3.view(out3.size(0),-1)out3 = self.attr3_layer2(out3)out_list.append(out3)#out4out4 = self.attr4_layer1(x)out4 = out4.view(out4.size(0),-1)out4 = self.attr4_layer2(out4)out_list.append(out4)#out5out5 = self.attr5_layer1(x)out5 = out5.view(out5.size(0),-1)out5 = self.attr5_layer2(out5)out_list.append(out5)#out6out6 = self.attr6_layer1(x)out6 = out6.view(out6.size(0),-1)out6 = self.attr6_layer2(out6)out_list.append(out6)#out7out7 = self.attr7_layer1(x)out7 = out7.view(out7.size(0),-1)out7 = self.attr7_layer2(out7)out_list.append(out7)#out8out8 = self.attr8_layer1(x)out8 = out8.view(out8.size(0),-1)out8 = self.attr8_layer2(out8)out_list.append(out8)#out9out9 = self.attr9_layer1(x)out9 = out9.view(out9.size(0),-1)out9 = self.attr9_layer2(out9)out_list.append(out9)#out10out10 = self.attr10_layer1(x)out10 = out10.view(out10.size(0),-1)out10 = self.attr10_layer2(out10)out_list.append(out10)#out11out11 = self.attr11_layer1(x)out11 = out11.view(out11.size(0),-1)out11 = self.attr11_layer2(out11)out_list.append(out11)#out12out12 = self.attr12_layer1(x)out12 = out12.view(out12.size(0),-1)out12 = self.attr12_layer2(out12)out_list.append(out12)#out13out13 = self.attr13_layer1(x)out13 = out13.view(out13.size(0),-1)out13 = self.attr13_layer2(out13)out_list.append(out13)#out14out14 = self.attr14_layer1(x)out14 = out14.view(out14.size(0),-1)out14 = self.attr14_layer2(out14)out_list.append(out14)#out15out15 = self.attr15_layer1(x)out15 = out15.view(out15.size(0),-1)out15 = self.attr15_layer2(out15)out_list.append(out15)#out16out16 = self.attr16_layer1(x)out16 = out16.view(out16.size(0),-1)out16 = self.attr16_layer2(out16)out_list.append(out16)#out17out17 = self.attr17_layer1(x)out17 = out17.view(out17.size(0),-1)out17 = self.attr17_layer2(out17)out_list.append(out17)#out18out18 = self.attr18_layer1(x)out18 = out18.view(out18.size(0),-1)out18 = self.attr18_layer2(out18)out_list.append(out18)#out19out19 = self.attr19_layer1(x)out19 = out19.view(out19.size(0),-1)out19 = self.attr19_layer2(out19)out_list.append(out19)#out20out20 = self.attr20_layer1(x)out20 = out20.view(out20.size(0),-1)out20 = self.attr20_layer2(out20)out_list.append(out20)#out21out21 = self.attr21_layer1(x)out21 = out21.view(out21.size(0),-1)out21 = self.attr21_layer2(out21)out_list.append(out21)#out22out22 = self.attr22_layer1(x)out22 = out22.view(out22.size(0),-1)out22 = self.attr22_layer2(out22)out_list.append(out22)#out23out23 = self.attr23_layer1(x)out23 = out23.view(out23.size(0),-1)out23 = self.attr23_layer2(out23)out_list.append(out23)#out24out24 = self.attr24_layer1(x)out24 = out24.view(out24.size(0),-1)out24 = self.attr24_layer2(out24)out_list.append(out24)#out25out25 = self.attr25_layer1(x)out25 = out25.view(out25.size(0),-1)out25 = self.attr25_layer2(out25)out_list.append(out25)#out26out26 = self.attr26_layer1(x)out26 = out26.view(out26.size(0),-1)out26 = self.attr26_layer2(out26)out_list.append(out26)#out27out27 = self.attr27_layer1(x)out27 = out27.view(out27.size(0),-1)out27 = self.attr27_layer2(out27)out_list.append(out27)#out28out28 = self.attr28_layer1(x)out28 = out28.view(out28.size(0),-1)out28 = self.attr28_layer2(out28)out_list.append(out28)#out29out29 = self.attr29_layer1(x)out29 = out29.view(out29.size(0),-1)out29 = self.attr29_layer2(out29)out_list.append(out29)#out30out30 = self.attr30_layer1(x)out30 = out30.view(out30.size(0),-1)out30 = self.attr30_layer2(out30)out_list.append(out30)#out31out31 = self.attr31_layer1(x)out31 = out31.view(out31.size(0),-1)out31 = self.attr31_layer2(out31)out_list.append(out31)#out32out32 = self.attr32_layer1(x)out32 = out32.view(out32.size(0),-1)out32 = self.attr32_layer2(out32)out_list.append(out32)#out33out33 = self.attr33_layer1(x)out33 = out33.view(out33.size(0),-1)out33 = self.attr33_layer2(out33)out_list.append(out33)#out34out34 = self.attr34_layer1(x)out34 = out34.view(out34.size(0),-1)out34 = self.attr34_layer2(out34)out_list.append(out34)#out35out35 = self.attr35_layer1(x)out35 = out35.view(out35.size(0),-1)out35 = self.attr35_layer2(out35)out_list.append(out35)#out36out36 = self.attr36_layer1(x)out36 = out36.view(out36.size(0),-1)out36 = self.attr36_layer2(out36)out_list.append(out36)#out37out37 = self.attr37_layer1(x)out37 = out37.view(out37.size(0),-1)out37 = self.attr37_layer2(out37)out_list.append(out37)#out38out38 = self.attr38_layer1(x)out38 = out38.view(out38.size(0),-1)out38 = self.attr38_layer2(out38)out_list.append(out38)#out39out39 = self.attr39_layer1(x)out39 = out39.view(out39.size(0),-1)out39 = self.attr39_layer2(out39)out_list.append(out39)return out_listmodule = face_attr()
#print(module)optimizer = optim.Adam(module.parameters(),lr = 0.001,weight_decay=1e-8)loss_list = []
for i in range(40):loss_func = nn.CrossEntropyLoss()loss_list.append(loss_func)
#loss_func = nn.CrossEntropyLoss()
for Epoch in range(50):all_correct_num = 0for ii,(img,label) in enumerate(train_dataloader):img = Variable(img)label = Variable(label)#    optimizer.zero_grad()output = module(img)optimizer.zero_grad()for i in range(40):loss = loss_list[i](output[i],label[:,i])loss.backward()_,predict = torch.max(output[i],1)correct_num = sum(predict==label[:,i])all_correct_num += correct_num.data[0]optimizer.step()        Accuracy =  all_correct_num *1.0/(len(train_dataset)*40.0)print('Epoch ={0},all_correct_num={1},Accuracy={2}'.format(Epoch,all_correct_num,Accuracy))torch.save(module,'W:/pic_data/face/CelebA/Img/face_attr40dro1.pkl')
'''
test_txt = 'W:/pic_data/face/CelebA/Img/test1000.txt'
test_dataset = myDataset(img_dir=img_root,img_txt = test_txt,transform= transform)
test_dataloader = Data.DataLoader(test_dataset,batch_size = batch_size,shuffle=True)module = torch.load('W:/pic_data/face/CelebA/Img/face_attr40dro1.pkl')
module.eval()
all_correct_num = 0
for ii,(img,label) in enumerate(test_dataloader):img = Variable(img)label = Variable(label)output = module(img)for i in range(40):_,predict = torch.max(output[i],1)correct_num = sum(predict==label[:,i])all_correct_num += correct_num.data[0]
Accuracy =  all_correct_num *1.0/(len(test_dataset)*40.0)
print('all_correct_num={0},Accuracy={1}'.format(all_correct_num,Accuracy))
'''

欢迎分享和留言。

分享请注明出处。

Pytorch实现人脸多属性识别相关推荐

pytorch 实现人脸检测与识别
pytorch + opencv 实现人脸检测与识别准备工作人脸检测 opencv实现人脸检测卷积神经网络 CNN 实现人脸检测数据导入 CNN模型训练人脸检测存在的问题人脸识别获取数 ...
pytorch实现人脸表情识别
数据来源CalebA人脸数据集(官网链接) 可以参照https://blog.csdn.net/think_three/article/details/80784952 1图片处理 2构建网络 3训练 ...
pytorch实现人脸识别_一步一步带你完成深度学习与对象检测之人脸识别
前期文章我们分享了opencv的人脸检测人工智能-OpenCV+Python实现人脸识别以及dlib的人脸检测与人脸识别人工智能-Dlib+Python实现人脸识别通过往期的分享,我们了解到人 ...
【计算视觉】人脸属性识别算法 | 性别+种族+年龄+表情
随着社会的发展,快速有效的自动身份验证在安防领域变的越来越迫切.由于生物特性是人的内在属性,具有很强的自身稳定性和个体差异性,是身份验证的最理想依据.利用人脸属性进行身份验证又是最自然直接的手段,相比 ...
基于PyTorch的人脸识别框架（翻译）
内容来源:https://github.com/XiaohangZhan/face_recognition_framework 基于PyTorch的人脸识别框架 Face recognition fr ...
python识别人脸多种属性_人脸检测及识别python实现系列（4）——卷积神经网络（CNN）入门...
人脸检测及识别python实现系列(4)--卷积神经网络(CNN)入门上篇博文我们准备好了2000张训练数据,接下来的几节我们将详细讲述如何利用这些数据训练我们的识别模型.前面说过,原博文给出的训练 ...
python识别人脸多种属性_9种人脸情绪识别、22类人脸属性、 15类皮肤质量识别，竹间智能要让情感机器人更理解人...
撰文 | 邱陆陆编辑 | 藤子「我们希望做情感机器人,希望透过机器更智能地理解一个人.」随着计算能力的提升.神经网络的步步深入,配以图像识别.语音识别的人工智能已经逐渐变得能够「听见」.「看见」. ...
行人属性数据集pa100k_基于InceptionV3的多数据集联合训练的行人外观属性识别方法与流程...
本发明涉及模式识别技术.智能监控技术等领域,具体的说,是基于Inception V3的多数据集联合训练的行人外观属性识别方法. 背景技术: 近年来,视频监控系统已经被广泛应用于安防领域.安防人员通过合 ...
人脸验证与识别——从模型训练到项目部署
前言 1.人脸验证其实是人脸识别中的一种,人脸验证要做的是1对1的验证,算法的验证模式是对当前人脸与另一张人脸做比对,然后给出得分值,可以按得分值来证明可以当前的人脸是否与另一给脸匹配上.这种使用最多 ...

Pytorch实现人脸多属性识别

Pytorch实现人脸多属性识别相关推荐

最新文章

热门文章