美文网首页大数据 爬虫Python AI SqlPython小哥哥
用AI人脸识别技术实现抖音特效 ！基本什么功能都能用Python

用AI人脸识别技术实现抖音特效 ！基本什么功能都能用Python

作者: 14e61d025165 | 来源:发表于2019-03-22 15:12 被阅读2次
<tt-image data-tteditor-tag="tteditorTag" contenteditable="false" class="syl1553238613120" data-render-status="finished" data-syl-blot="image" style="box-sizing: border-box; cursor: text; color: rgb(34, 34, 34); font-family: "PingFang SC", "Hiragino Sans GB", "Microsoft YaHei", "WenQuanYi Micro Hei", "Helvetica Neue", Arial, sans-serif; font-size: 16px; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-align: left; text-indent: 0px; text-transform: none; white-space: pre-wrap; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; background-color: rgb(255, 255, 255); text-decoration-style: initial; text-decoration-color: initial; display: block;"> image

<input class="pgc-img-caption-ipt" placeholder="图片描述(最多50字)" value="" style="box-sizing: border-box; outline: 0px; color: rgb(102, 102, 102); position: absolute; left: 187.5px; transform: translateX(-50%); padding: 6px 7px; max-width: 100%; width: 375px; text-align: center; cursor: text; font-size: 12px; line-height: 1.5; background-color: rgb(255, 255, 255); background-image: none; border: 0px solid rgb(217, 217, 217); border-radius: 4px; transition: all 0.2s cubic-bezier(0.645, 0.045, 0.355, 1);"></tt-image>

导言
欢迎大家加入小编创建的Python行业交流群，有大牛答疑，有资源共享，有企业招人！是一个非常不错的交流基地！群号：683380553

  • 为什么会写这个人脸例子？

<bi style="box-sizing: border-box; display: block;">浏览博客的过程，无意发现了一篇名为deal-with-it-generator-face-recognition的文章，通过这篇文章，使我有了写这个例子的想法，尤其是现在很多短视频APP中经常出现这样的效果，感觉还是有点好玩的。</bi>

私信小编001 有神秘惊喜大礼包！

  • 感谢!

<bi style="box-sizing: border-box; display: block;">写这个例子初衷与资料来自burningion的分享。</bi>

  • 变化？

<bi style="box-sizing: border-box; display: block;">deal-with-it-generator-face-recognition 这篇文章是一个戴眼镜的简单例子及构思，我在其原基础上，添加了烟卷的部分，并且把代码结构重构了一下，使其更易拓展和维护，也易于阅读。</bi>

实现流程

<bi style="box-sizing: border-box; display: block;">程序从命令行参数获取图片信息，然后，它将使用Dlib中的人脸检测算法来查看是否有人脸存在。如果有，它将为每个人脸创建一个结束位置，眼镜和烟卷会移动到那里结束。</bi><bi style="box-sizing: border-box; display: block;">然后我们需要缩放和旋转我们的眼镜以适合每个人的脸。我们将使用从Dlib的68点模型返回的点集来找到眼睛的中心，并为它们之间的空间旋转。</bi><bi style="box-sizing: border-box; display: block;">在我们找到眼镜的最终位置和旋转后，我们可以为gif制作动画，眼镜从屏幕顶部进入。我们将使用MoviePy和一个make_frame函数绘制它。</bi><bi style="box-sizing: border-box; display: block;">同理烟卷也是这样。</bi><bi style="box-sizing: border-box; display: block;">应用程序的体系结构非常简单。我们首先接收图片，然后将其转换为灰度NumPy数组。假如没有人脸，程序会自己退出，如果存在，我们就可以将检测到的人脸信息传递到人脸方向预测模型中。</bi><bi style="box-sizing: border-box; display: block;">通过返回的脸部方向，我们可以选择眼睛，缩放和旋转我们的眼镜框架以适合人的面部大小。</bi><bi style="box-sizing: border-box; display: block;">当然这个程序不仅仅只针对于一张人脸，可以检测多个人脸信息。</bi><bi style="box-sizing: border-box; display: block;">最后，通过获取的人脸列表，我们可以使用MoviePy创建一个绘图，然后生成我们的动画gif。</bi>

  • 导入对应的工具包

<pre spellcheck="false" style="box-sizing: border-box; margin: 5px 0px; padding: 5px 10px; border: 0px; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-variant-numeric: inherit; font-variant-east-asian: inherit; font-weight: 400; font-stretch: inherit; font-size: 16px; line-height: inherit; font-family: inherit; vertical-align: baseline; cursor: text; counter-reset: list-1 0 list-2 0 list-3 0 list-4 0 list-5 0 list-6 0 list-7 0 list-8 0 list-9 0; background-color: rgb(240, 240, 240); border-radius: 3px; white-space: pre-wrap; color: rgb(34, 34, 34); letter-spacing: normal; orphans: 2; text-align: left; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;">import moviepy.editor as mpy
import numpy as np
from PIL import Image
from imutils import face_utils
try:
from dlib import get_frontal_face_detector, shape_predictor
except ImportError:
raise
</pre>

  • 创建人脸识别的工具类FaceDetect及其对应的方法

<pre spellcheck="false" style="box-sizing: border-box; margin: 5px 0px; padding: 5px 10px; border: 0px; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-variant-numeric: inherit; font-variant-east-asian: inherit; font-weight: 400; font-stretch: inherit; font-size: 16px; line-height: inherit; font-family: inherit; vertical-align: baseline; cursor: text; counter-reset: list-1 0 list-2 0 list-3 0 list-4 0 list-5 0 list-6 0 list-7 0 list-8 0 list-9 0; background-color: rgb(240, 240, 240); border-radius: 3px; white-space: pre-wrap; color: rgb(34, 34, 34); letter-spacing: normal; orphans: 2; text-align: left; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;">class FaceDetect(object):
pass
</pre>

  • 创建detector及predictor两个属性，用来加载dlib库函数

<pre spellcheck="false" style="box-sizing: border-box; margin: 5px 0px; padding: 5px 10px; border: 0px; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-variant-numeric: inherit; font-variant-east-asian: inherit; font-weight: 400; font-stretch: inherit; font-size: 16px; line-height: inherit; font-family: inherit; vertical-align: baseline; cursor: text; counter-reset: list-1 0 list-2 0 list-3 0 list-4 0 list-5 0 list-6 0 list-7 0 list-8 0 list-9 0; background-color: rgb(240, 240, 240); border-radius: 3px; white-space: pre-wrap; color: rgb(34, 34, 34); letter-spacing: normal; orphans: 2; text-align: left; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;">@property
def detector(self):
"""
检测是否有人脸
:return:
"""
return get_frontal_face_detector()
@property
def predictor(self):
"""
预测人脸方向
:return:
"""
return shape_predictor('shape_predictor_68_face_landmarks.dat')
</pre>

  • 创建init_mask函数，用来加载面具信息(墨镜，烟卷等信息)

<pre spellcheck="false" style="box-sizing: border-box; margin: 5px 0px; padding: 5px 10px; border: 0px; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-variant-numeric: inherit; font-variant-east-asian: inherit; font-weight: 400; font-stretch: inherit; font-size: 16px; line-height: inherit; font-family: inherit; vertical-align: baseline; cursor: text; counter-reset: list-1 0 list-2 0 list-3 0 list-4 0 list-5 0 list-6 0 list-7 0 list-8 0 list-9 0; background-color: rgb(240, 240, 240); border-radius: 3px; white-space: pre-wrap; color: rgb(34, 34, 34); letter-spacing: normal; orphans: 2; text-align: left; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;">@classmethod
def load(cls, img_src):
"""
加载图片转为Image对象
:param img_src:
:return:
"""
return Image.open(img_src)
def init_mask(self):
"""
加载面具
:return:
"""
self.deal, self.text, self.cigarette = (
self.load(x) for x in ["../images/deals.png", "../images/text.png", "../images/cigarette.png"]
)
</pre>

  • 创建收集人脸信息的对应方法

<bi style="box-sizing: border-box; display: block;">首先get_glasses_info方法会根据当前人脸的特征值及图片基础设置，对图片中人脸进行面部定位，计算眼角倾斜度，来改变眼镜最终位置及角度,并将此信息返回给面部定位函数</bi><bi style="box-sizing: border-box; display: block;">get_cigarette_info 方法会根据当前人脸的特征值及图片基础设置，来计算人脸嘴巴的位置，并将其返回给面部定位函数。</bi><bi style="box-sizing: border-box; display: block;">orientation 方法会将基础的人脸信息通过"get_cigarette_info"和"get_glasses_info"方法处理后，再一并返回给画图函数，供其画图。</bi>

<pre spellcheck="false" style="box-sizing: border-box; margin: 5px 0px; padding: 5px 10px; border: 0px; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-variant-numeric: inherit; font-variant-east-asian: inherit; font-weight: 400; font-stretch: inherit; font-size: 16px; line-height: inherit; font-family: inherit; vertical-align: baseline; cursor: text; counter-reset: list-1 0 list-2 0 list-3 0 list-4 0 list-5 0 list-6 0 list-7 0 list-8 0 list-9 0; background-color: rgb(240, 240, 240); border-radius: 3px; white-space: pre-wrap; color: rgb(34, 34, 34); letter-spacing: normal; orphans: 2; text-align: left; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;">def get_glasses_info(self, face_shape, face_width):
"""
获取当前面部的眼镜信息
:param face_shape:
:param face_width:
:return:
"""
left_eye = face_shape[36:42]
right_eye = face_shape[42:48]
left_eye_center = left_eye.mean(axis=0).astype("int")
right_eye_center = right_eye.mean(axis=0).astype("int")
y = left_eye_center[1] - right_eye_center[1]
x = left_eye_center[0] - right_eye_center[0]
eye_angle = np.rad2deg(np.arctan2(y, x))
deal = self.deal.resize(
(face_width, int(face_width * self.deal.size[1] / self.deal.size[0])),
resample=Image.LANCZOS)
deal = deal.rotate(eye_angle, expand=True)
deal = deal.transpose(Image.FLIP_TOP_BOTTOM)
left_eye_x = left_eye[0, 0] - face_width // 4
left_eye_y = left_eye[0, 1] - face_width // 6
return {"image": deal, "pos": (left_eye_x, left_eye_y)}
def get_cigarette_info(self, face_shape, face_width):
"""
获取当前面部的眼镜信息
:param face_shape:
:param face_width:
:return:
"""
mouth = face_shape[49:68]
mouth_center = mouth.mean(axis=0).astype("int")
cigarette = self.cigarette.resize(
(face_width, int(face_width * self.cigarette.size[1] / self.cigarette.size[0])),
resample=Image.LANCZOS)
x = mouth[0, 0] - face_width + int(16 * face_width / self.cigarette.size[0])
y = mouth_center[1]
return {"image": cigarette, "pos": (x, y)}
def orientation(self):
"""
人脸定位
:return:
"""
faces = []
for rect in self.rects:
face = {}
face_shades_width = rect.right() - rect.left()
predictor_shape = self.predictor(self.img_gray, rect)
face_shape = face_utils.shape_to_np(predictor_shape)
face['cigarette'] = self.get_cigarette_info(face_shape, face_shades_width)
face['glasses'] = self.get_glasses_info(face_shape, face_shades_width)
faces.append(face)
return faces
</pre>

  • 那我们开始来实现画图函数drawing

<bi style="box-sizing: border-box; display: block;">根据传入的参数t来计算生成GIF的进度,这里设置画图周期前2秒，来移动面具（即眼镜和烟卷），在两秒前结束移动，然后再画出字体，基本就是这个流程。</bi><bi style="box-sizing: border-box; display: block;">面具移动的实现就是来动态更新面具的纵坐标。</bi>

<pre spellcheck="false" style="box-sizing: border-box; margin: 5px 0px; padding: 5px 10px; border: 0px; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-variant-numeric: inherit; font-variant-east-asian: inherit; font-weight: 400; font-stretch: inherit; font-size: 16px; line-height: inherit; font-family: inherit; vertical-align: baseline; cursor: text; counter-reset: list-1 0 list-2 0 list-3 0 list-4 0 list-5 0 list-6 0 list-7 0 list-8 0 list-9 0; background-color: rgb(240, 240, 240); border-radius: 3px; white-space: pre-wrap; color: rgb(34, 34, 34); letter-spacing: normal; orphans: 2; text-align: left; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;">def drawing(self, t):
"""
动态画图
:param t:
:return:
"""
draw_img = self.image.convert('RGBA')
if t == 0:
return np.asarray(draw_img)
for face in self.orientation():
if t <= self.duration - 2:
current_x = int(face["glasses"]["pos"][0])
current_y = int(face["glasses"]["pos"][1] * t / (self.duration - 2))
draw_img.paste(face["glasses"]["image"], (current_x, current_y), face["glasses"]["image"])
cigarette_x = int(face["cigarette"]["pos"][0])
cigarette_y = int(face["cigarette"]["pos"][1] * t / (self.duration - 2))
draw_img.paste(face["cigarette"]["image"], (cigarette_x, cigarette_y), face["cigarette"]["image"])
else:
draw_img.paste(face["glasses"]["image"], face["glasses"]["pos"], face["glasses"]["image"])
draw_img.paste(face["cigarette"]["image"], face["cigarette"]["pos"], face["cigarette"]["image"])
draw_img.paste(self.text, (75, draw_img.height // 2 + 128), self.text)
return np.asarray(draw_img)
</pre>

  • 以上基本的函数及实现已经完成，让我们来设置一下初始化参数

<pre spellcheck="false" style="box-sizing: border-box; margin: 5px 0px; padding: 5px 10px; border: 0px; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-variant-numeric: inherit; font-variant-east-asian: inherit; font-weight: 400; font-stretch: inherit; font-size: 16px; line-height: inherit; font-family: inherit; vertical-align: baseline; cursor: text; counter-reset: list-1 0 list-2 0 list-3 0 list-4 0 list-5 0 list-6 0 list-7 0 list-8 0 list-9 0; background-color: rgb(240, 240, 240); border-radius: 3px; white-space: pre-wrap; color: rgb(34, 34, 34); letter-spacing: normal; orphans: 2; text-align: left; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;">class FaceDetect(object):
def init(self, img_src, gif_path=None):
self.gif_max_width = 500
self.duration = 4
self.image = self.load(img_src).convert('RGBA')
self.img_gray = None
self.rects = None
self.deal = None
self.text = None
self.cigarette = None
if not self.validate:
print("没有检测到人脸，程序退出.")
exit(1)
self.init_mask()
self.make_gif(gif_path=gif_path)
@property
def validate(self):
"""
验证是否存在人脸，如果不存在返回False
:return:
"""
if self.image.size[0] > self.gif_max_width:
scaled_height = int(self.gif_max_width * self.image.size[1] / self.image.size[0])
self.image.thumbnail((self.gif_max_width, scaled_height))
self.img_gray = np.array(self.image.convert('L'))
self.rects = self.detector(self.img_gray, 0)
return len(self.rects) > 0
def make_gif(self, gif_path=None):
"""
:param gif_path: 保存路径
:return:
"""
gif_path = gif_path or "deal.gif"
animation = mpy.VideoClip(self.drawing, duration=self.duration)
animation.write_gif(gif_path, fps=self.duration)
</pre>

  • 最后我们实现一下main函数

<pre spellcheck="false" style="box-sizing: border-box; margin: 5px 0px; padding: 5px 10px; border: 0px; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-variant-numeric: inherit; font-variant-east-asian: inherit; font-weight: 400; font-stretch: inherit; font-size: 16px; line-height: inherit; font-family: inherit; vertical-align: baseline; cursor: text; counter-reset: list-1 0 list-2 0 list-3 0 list-4 0 list-5 0 list-6 0 list-7 0 list-8 0 list-9 0; background-color: rgb(240, 240, 240); border-radius: 3px; white-space: pre-wrap; color: rgb(34, 34, 34); letter-spacing: normal; orphans: 2; text-align: left; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;">if name == 'main':
# 运行 python input_static_pic_to_gif2_for_class.py -image ../images/1.jpg
import argparse
parser = argparse.ArgumentParser()
parser.add_argument("-image", required=True, help="path to input image")
parser.add_argument("-save", required=False, default="deal.gif", help="path to output image")
args = parser.parse_args()
FaceDetect(args.image, args.save)
</pre>

  • 写到这里，这个小功能就已经实现了，大家不妨来使用一下

写在最后

相关文章

网友评论

    本文标题：用AI人脸识别技术实现抖音特效 ！基本什么功能都能用Python

    本文链接：https://www.haomeiwen.com/subject/lgfrvqtx.html

    延伸阅读

    深度阅读

    • 您也可以注册成为美文阅读网的作者，发表您的原创作品、分享您的心情！

    栏目导航

    热点阅读

    大数据 爬虫Python AI Sql

    Python小哥哥

    关于我们|服务条款|联系我们|用AI人脸识别技术实现抖音特效 ！基本什么功能都能用Python|投稿指南|网站地图|RSS订阅|排版工具|手机版

    提供经典美文摘抄,优美散文欣赏,现代诗歌精选,短篇小说,心情随笔,表白情书范文,故事会在线阅读欣赏

    Copyright © 2014-2023 Haomeiwen.com All Rights Reserved. 好美文阅读网 版权所有

    备案信息：桂公网安备 45052102000051号 · 桂ICP备13007215号-3

    本站所收录作品、热点评论等信息部分来源互联网，目的只是为了系统归纳学习和传递资讯

    所有作品版权归原创作者所有，与本站立场无关，如不慎侵犯了你的权益，请联系我们告知，我们将做删除处理！