C++ Dlib人脸地标检测链接器错误

我最近从源代码安装了OpenCV 3.0。然后我在macOS Sierra 10.12上通过自制安装了Dlib，因为我想了解更多关于计算机视觉的知识

我试图编译在Dlib的网站上找到的程序，我得到了以下链接器错误，我一直没有找到正确的编译器标志来修复这个错误。我也不确定我是否真的安装了所需的软件

我收到的链接器错误是：

Undefined symbols for architecture x86_64:

"_DGifCloseFile", referenced from:
void dlib::load_image<dlib::array2d<dlib::rgb_pixel, dlib::memory_manager_stateless_kernel_1<char> > >(dlib::array2d<dlib::rgb_pixel, dlib::memory_manager_stateless_kernel_1<char> >&, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&) in face_landmark_detection-7a9d91.o
"_DGifOpenFileName", referenced from:
  void dlib::load_image<dlib::array2d<dlib::rgb_pixel, dlib::memory_manager_stateless_kernel_1<char> > >(dlib::array2d<dlib::rgb_pixel, dlib::memory_manager_stateless_kernel_1<char> >&, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&) in face_landmark_detection-7a9d91.o
"_DGifSlurp", referenced from:
  void dlib::load_image<dlib::array2d<dlib::rgb_pixel, dlib::memory_manager_stateless_kernel_1<char> > >(dlib::array2d<dlib::rgb_pixel, dlib::memory_manager_stateless_kernel_1<char> >&, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&) in face_landmark_detection-7a9d91.o
"_cblas_dgemm", referenced from:
  dlib::blas_bindings::cblas_gemm(dlib::blas_bindings::CBLAS_ORDER, dlib::blas_bindings::CBLAS_TRANSPOSE, dlib::blas_bindings::CBLAS_TRANSPOSE, int, int, int, double, double const*, int, double const*, int, double, double*, int) in face_landmark_detection-7a9d91.o
"_cblas_saxpy", referenced from:
  dlib::blas_bindings::cblas_axpy(int, float, float const*, int, float*, int) in face_landmark_detection-7a9d91.o
"_cblas_sscal", referenced from:
  dlib::blas_bindings::cblas_scal(int, float, float*) in face_landmark_detection-7a9d91.o
"_dgesvd_", referenced from:
  dlib::lapack::binding::gesvd(char, char, int, int, double*, int, double*, double*, int, double*, int, double*, int) in face_landmark_detection-7a9d91.o
ld: symbol(s) not found for architecture x86_64

架构x86_64的未定义符号： “_DGifCloseFile”，引用自： void dlib：：加载面中的图像（dlib：：array2d&，std:uuu 1：：基本字符串常量&）u landmark_detection-7a9d91.o “DGifOpenFileName”，引用自： void dlib：：加载面中的图像（dlib：：array2d&，std:uuu 1：：基本字符串常量&）u landmark_detection-7a9d91.o “_DGifSlurp”，引用自： void dlib：：加载面中的图像（dlib：：array2d&，std:uuu 1：：基本字符串常量&）u landmark_detection-7a9d91.o “_cblas_dgemm”，引用自： dlib:：blas_bindings:：cblas_gemm（dlib:：blas_bindings:：cblas_ORDER，dlib:：blas_bindings:：cblas_TRANSPOSE，dlib:：blas_bindings:：cblas_TRANSPOSE，int，int，int，double，double const*，int，double，double*，int）在face_landmark\u detection-7a9d91.o中 “\u cblas\u saxpy”，引用自： dlib:：blas_bindings:：cblas_axpy（int，float，float const*，int，float*，int）in face_landmark_detection-7a9d91.o “\u cblas\u sscal”，引用自： dlib:：blas_bindings:：cblas_scal（int，float，float*）在脸上_landmark_detection-7a9d91.o “_dgesvd_”，引用自： dlib:：lapack:：binding:：gesvd（字符、字符、int、int、double*、int、double*、int、double*、int、double*、int、int、double*、int）中的面标记检测-7a9d91.o ld:找不到架构x86_64的符号 叮当声：错误：链接器命令失败，退出代码为1（使用-v查看调用）

（我安装了--HEAD，因为

brew安装opencv3

对我不起作用）
自制安装OpenCV:

brew点击自制/科学；brew安装--头部opencv3

自制安装的Dlib:

brew安装Dlib

编译标志：

g++-std=c++11 face\u landmark\u detection.cpp'pkg config--cflags--libs opencv dlib-1'

pkg config--cflags--libs opencv dlib-1的输出：

-I/usr/local/Cellar/opencv3/HEAD-de35c59_4/include/opencv-I/usr/local/Cellar/opencv3/HEAD-de35c59_4/include-I/usr/local/Cellar/dlib/19.1_2/include-I/usr/local/Cellar/libpng/1.6.25/include/L/usr/local/Cellar/opencv3/HEAD-de35c59_4/lib-L-L/L-L/L/L/L/L/Cellar/dlib/19.1/L/L/L/L/L/L/L/dlib/L/L/L/L/L/L/L/L/L/L/L/L/L/L/L/L/L/L/-lopencv_缝合-lopencv_对象检测-lopencv_superres-lopencv_videostab-lopencv_calib3d-lopencv_功能2d-lopencv_highgui-lopencv_视频-lopencv_图片-lopencv_ml-lopencv_imgproc-lopencv_flann-lopencv_核心-ldlib-lpng16

---

有人能告诉我哪些文件需要链接吗？

使用cmake如何？您使用的是什么操作系统？哦！我忘了添加操作系统。我在第一句话中加了它。我在OSX El Capitan上。我想用自制软件安装它们，这样如果任何一个库发生任何更新，我的工作量就会减少。如果我不明白这一点，我完全会用cmake安装这两个软件。使用cmake怎么样？你使用的是什么操作系统？哦！我忘了添加操作系统。我在第一句话中加了它。我在OSX El Capitan上。我想用自制软件安装它们，这样如果任何一个库发生任何更新，我的工作量就会减少。如果我不明白这一点，我完全会用cmake安装这两个。