基于TensorFlow和FaceNet的人脸检测的学习

57 阅读 0 评论 38 点赞

我是靠谱客的博主单身人生，最近开发中收集的这篇文章主要介绍基于TensorFlow和FaceNet的人脸检测的学习，觉得挺不错的，现在分享给大家，希望可以做个参考。

概述

参考书《TensorFlow技术解析与实践》

做这个实验之前默认你的TensorFlow环境已经搭建好。

下载FaceNet源码

：git clone https://github.com/davidsandberg/facenet.git

下载LFW数据集

点击网址下载：http://vis-www.cs.umass.edu/lfw/lfw.tgz
mkdir -p $YOURHOME/facenet/datasets/lfw/raw,这个指令是在你下载的facenet所在文件夹下新建datasets/lfw/raw这几个文件夹，然后将你下载的LFW数据集解压到raw这个文件夹下。

数据预处理

：export PYTHON=$YOURHOME/facenet/src
：cd $YOURHOME/facenet
：python src/align/align_dataset_mtcnn.py $YOURHOME/facenet/datasets/lfw/raw $YOURHOME/facenet/datasets/lfw/lfw_mtcnnpy_160 --image_size 160 --margin 32 --random_order--gpu_memory_fraction 0.25
这需要一点时间。接下来下载预训练模型，百度网盘。链接: https://pan.baidu.com/s/1HOqQ8lGdQUpYFKNJiYN_6Q 提取码: ai5u
下载完成后：mkdir -p $YOURHOME/facenet/models/facenet，将下载的20170512-110547.zip解压到刚刚创建的这个文件夹。

进行检测

：cd $YOURHOME/facenet
：python src/validate_on_lfw.py $YOURHOME/facenet/datasets/lfw/lfw_mtcnnpy_160 models/facenet/20170512-110547
这里正常步骤就已经完成，下面是我遇到的一些问题。

结果比对

请查看facenet/data/pairs.txt的数据，匹配的人名和编号如下：
A 1 4，表示A的第一张和第四张是一个人。相反另一种表示就是不匹配的。

遇到的问题

问题一：ValueError: Input 0 of node Reshape was passed int32 from batch_join:1 incompatible with expected int64.
解决方法：打开validate_on_lfw.py，找到这三个地方，data_flow_ops.FIFOQueue，labels_placeholder，control_placeholder，将他们的tf.int32全部换成tf.int64重新运行即可。
其它问题：在执行的之后有到一两个问题，但是发现我解压的20170512-110547文件夹的内容被不明力量的给修改了，你可以检查一下自己的，然后重新解压即可。

validate_on_lfw.py流程解析

源代码如下：

"""Validate a face recognizer on the "Labeled Faces in the Wild" dataset (http://vis-www.cs.umass.edu/lfw/).
Embeddings are calculated using the pairs from http://vis-www.cs.umass.edu/lfw/pairs.txt and the ROC curve
is calculated and plotted. Both the model metagraph and the model parameters need to exist
in the same directory, and the metagraph should have the extension '.meta'.
"""
# MIT License
# 
# Copyright (c) 2016 David Sandberg
# 
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
# 
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
# 
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

import tensorflow as tf
import numpy as np
import argparse
import facenet
import lfw
import os
import sys
from tensorflow.python.ops import data_flow_ops
from sklearn import metrics
from scipy.optimize import brentq
from scipy import interpolate

def main(args):
  
    with tf.Graph().as_default():
      
        with tf.Session() as sess:
            
            # Read the file containing the pairs used for testing
            # 读入pairs文件
            pairs = lfw.read_pairs(os.path.expanduser(args.lfw_pairs))

            # Get the paths for the corresponding images
            # 获取文件路径和是否匹配的关系对
            paths, actual_issame = lfw.get_paths(os.path.expanduser(args.lfw_dir), pairs)
            
            image_paths_placeholder = tf.placeholder(tf.string, shape=(None,1), name='image_paths')
            labels_placeholder = tf.placeholder(tf.int64, shape=(None,1), name='labels')
            batch_size_placeholder = tf.placeholder(tf.int32, name='batch_size')
            control_placeholder = tf.placeholder(tf.int64, shape=(None,1), name='control')
            phase_train_placeholder = tf.placeholder(tf.bool, name='phase_train')
 
            nrof_preprocess_threads = 4
            image_size = (args.image_size, args.image_size)
            eval_input_queue = data_flow_ops.FIFOQueue(capacity=2000000,
                                        dtypes=[tf.string, tf.int64, tf.int64],
                                        shapes=[(1,), (1,), (1,)],
                                        shared_name=None, name=None)
            eval_enqueue_op = eval_input_queue.enqueue_many([image_paths_placeholder, labels_placeholder, control_placeholder], name='eval_enqueue_op')
            image_batch, label_batch = facenet.create_input_pipeline(eval_input_queue, image_size, nrof_preprocess_threads, batch_size_placeholder)
     
            # Load the model
            # 加载模型
            input_map = {'image_batch': image_batch, 'label_batch': label_batch, 'phase_train': phase_train_placeholder}
            facenet.load_model(args.model, input_map=input_map)

            # Get output tensor
            # 获取输入输出的张量
            embeddings = tf.get_default_graph().get_tensor_by_name("embeddings:0")
#              
            coord = tf.train.Coordinator()
            tf.train.start_queue_runners(coord=coord, sess=sess)

            evaluate(sess, eval_enqueue_op, image_paths_placeholder, labels_placeholder, phase_train_placeholder, batch_size_placeholder, control_placeholder,
                embeddings, label_batch, paths, actual_issame, args.lfw_batch_size, args.lfw_nrof_folds, args.distance_metric, args.subtract_mean,
                args.use_flipped_images, args.use_fixed_image_standardization)

              
def evaluate(sess, enqueue_op, image_paths_placeholder, labels_placeholder, phase_train_placeholder, batch_size_placeholder, control_placeholder,
        embeddings, labels, image_paths, actual_issame, batch_size, nrof_folds, distance_metric, subtract_mean, use_flipped_images, use_fixed_image_standardization):
    # Run forward pass to calculate embeddings
    # 使用前向传播来验证
    print('Runnning forward pass on LFW images')
    
    # Enqueue one epoch of image paths and labels
    nrof_embeddings = len(actual_issame)*2  # nrof_pairs * nrof_images_per_pair
    nrof_flips = 2 if use_flipped_images else 1
    nrof_images = nrof_embeddings * nrof_flips
    labels_array = np.expand_dims(np.arange(0,nrof_images),1)
    image_paths_array = np.expand_dims(np.repeat(np.array(image_paths),nrof_flips),1)
    control_array = np.zeros_like(labels_array, np.int32)
    if use_fixed_image_standardization:
        control_array += np.ones_like(labels_array)*facenet.FIXED_STANDARDIZATION
    if use_flipped_images:
        # Flip every second image
        control_array += (labels_array % 2)*facenet.FLIP
    sess.run(enqueue_op, {image_paths_placeholder: image_paths_array, labels_placeholder: labels_array, control_placeholder: control_array})
    
    embedding_size = int(embeddings.get_shape()[1])
    assert nrof_images % batch_size == 0, 'The number of LFW images must be an integer multiple of the LFW batch size'
    nrof_batches = nrof_images // batch_size
    emb_array = np.zeros((nrof_images, embedding_size))
    lab_array = np.zeros((nrof_images,))
    for i in range(nrof_batches):
        feed_dict = {phase_train_placeholder:False, batch_size_placeholder:batch_size}
        emb, lab = sess.run([embeddings, labels], feed_dict=feed_dict)
        lab_array[lab] = lab
        emb_array[lab, :] = emb
        if i % 10 == 9:
            print('.', end='')
            sys.stdout.flush()
    print('')
    embeddings = np.zeros((nrof_embeddings, embedding_size*nrof_flips))
    if use_flipped_images:
        # Concatenate embeddings for flipped and non flipped version of the images
        embeddings[:,:embedding_size] = emb_array[0::2,:]
        embeddings[:,embedding_size:] = emb_array[1::2,:]
    else:
        embeddings = emb_array

    assert np.array_equal(lab_array, np.arange(nrof_images))==True, 'Wrong labels used for evaluation, possibly caused by training examples left in the input pipeline'
    # 使用十折交叉验证计算准确率和验证率
    tpr, fpr, accuracy, val, val_std, far = lfw.evaluate(embeddings, actual_issame, nrof_folds=nrof_folds, distance_metric=distance_metric, subtract_mean=subtract_mean)
    
    print('Accuracy: %2.5f+-%2.5f' % (np.mean(accuracy), np.std(accuracy)))
    print('Validation rate: %2.5f+-%2.5f @ FAR=%2.5f' % (val, val_std, far))
    
    # 得到auc和等错误率eer的值
    auc = metrics.auc(fpr, tpr)
    print('Area Under Curve (AUC): %1.3f' % auc)
    eer = brentq(lambda x: 1. - x - interpolate.interp1d(fpr, tpr)(x), 0., 1.)
    print('Equal Error Rate (EER): %1.3f' % eer)
    
def parse_arguments(argv):
    parser = argparse.ArgumentParser()
    
    parser.add_argument('lfw_dir', type=str,
        help='Path to the data directory containing aligned LFW face patches.')
    parser.add_argument('--lfw_batch_size', type=int,
        help='Number of images to process in a batch in the LFW test set.', default=100)
    parser.add_argument('model', type=str, 
        help='Could be either a directory containing the meta_file and ckpt_file or a model protobuf (.pb) file')
    parser.add_argument('--image_size', type=int,
        help='Image size (height, width) in pixels.', default=160)
    parser.add_argument('--lfw_pairs', type=str,
        help='The file containing the pairs to use for validation.', default='data/pairs.txt')
    parser.add_argument('--lfw_nrof_folds', type=int,
        help='Number of folds to use for cross validation. Mainly used for testing.', default=10)
    parser.add_argument('--distance_metric', type=int,
        help='Distance metric  0:euclidian, 1:cosine similarity.', default=0)
    parser.add_argument('--use_flipped_images', 
        help='Concatenates embeddings for the image and its horizontally flipped counterpart.', action='store_true')
    parser.add_argument('--subtract_mean', 
        help='Subtract feature mean before calculating distance.', action='store_true')
    parser.add_argument('--use_fixed_image_standardization', 
        help='Performs fixed standardization of images.', action='store_true')
    return parser.parse_args(argv)

if __name__ == '__main__':
    main(parse_arguments(sys.argv[1:]))