Run legacy FDDB Evaluation Scripts on macOS Mojave (10.14)

Background

FDDB, nearly 20 years old codes (even with a April 1, 2020 updated Gnuplot scripts for generating the ROC curves), should be run just fine - Not on the MAC!

Build and Errors

Build is easy, really.

• Choose Mac because only Makefile available with pkg-config for setting up OpenCV paths and libraries
• Should be easy to build on Windows anyway, either by manually specify the OpenCV path/libs, or create a simple CMakefile then build it

ERROR: Illegal instruction: 4

What is the “Illegal Instruction: 4”

The “illegal instruction” message is simply telling you that your binaries contain instructions the version of the OS that you are attempting to run them under does not understand. I can’t give you the precise meaning of 4 but I expect that is internal to Apple.

REASON

To do with the warning messages in below during compilation

RegionsSingleImage.cpp:38:2: warning: delete called on 'Region' that is abstract but has non-virtual destructor
      [-Wdelete-non-virtual-dtor]
        delete(list->at(i)); 
        ^
1 warning generated.

evaluate.cpp:320:5: warning: delete called on 'RegionsSingleImage' that is abstract but has non-virtual destructor
      [-Wdelete-non-virtual-dtor]
    delete(annot);
    ^
evaluate.cpp:321:5: warning: delete called on 'RegionsSingleImage' that is abstract but has non-virtual destructor
      [-Wdelete-non-virtual-dtor]
    delete(det);
    ^
2 warnings generated.

Interfaces in C++ (Abstract Classes

What does ‘has virtual method … but non-virtual destructor’ warning mean during C++ compilation?

If a class has a virtual method, that means you want other classes to inherit from it. These classes could be destroyed through a base-class-reference or pointer, but this would only work if the base-class has a virtual destructor. If you have a class that is supposed to be usable polymorphically, it should also be deletable polymorphically.

When should your destructor be virtual? [duplicate]

When delete a derived class object with no virtual destructor, it is undefined behavior , apparently modern macOS and build tools are more strict with the rules

浅谈 C++ 中的 new/delete 和 new[]/delete[]

Solution

add TWO ‘virtual’ keywords

• virtual ~Region(); in Region.hpp
• virtual ~RegionsSingleImage(); in RegionsSingleImage.hpp

Generate detection results

FDDB evaluation code

def get_img_relative_path():
    """
    :return: ['2002/08/11/big/img_344', '2002/08/02/big/img_473', ......]
    """
    f_name = 'E:/face_rec/face__det_rec_code/face_det/FDDB-folds/all_img_files.txt'
    lst_name = open(f_name).read().split('\n')

    return lst_name

def write_lines_to_txt(lst):
    # lst = ['line1', 'line2', 'line3']
    f_path = 'fddb_rect_ret.txt'
    with open(f_path, 'w') as fp:

        for line in lst:
            fp.write("%s\n" % line)

# For example use opencv to face detection
def detect_face_lst(img):
    """
    :param img: opencv image 
    :return: face rectangles [[x, y, w, h], ..........]
    """
    m_path = 'D:/opencv/sources/data/haarcascades/haarcascade_frontalface_default.xml'
    face_cascade = cv2.CascadeClassifier(m_path)

    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

    faces = face_cascade.detectMultiScale(gray, 1.3, 5)

    return faces


def generate_fddb_ret():
    # The directory from which we get the test images from FDDB
    img_base_dir = 'E:/face_rec/face__det_rec_code/face_det/originalPics/'

    # All the images relative path, like '['2002/08/11/big/img_344', '2002/08/02/big/img_473', ......]'
    lst_img_name = get_img_relative_path()

    # Store detect result, like:
    # ['2002/08/11/big/img_344', '1', '10 10 50 50 1', .............]
    lst_write2_fddb_ret = []

    try:
        for img_name in lst_img_name:
            img_full_name = img_base_dir + img_name + '.jpg'
            img = cv2.imread(img_full_name)

            if img == None:
                print 'error %s not exists, can not generate complete fddb evaluate file' % img_full_name
                return -1

            lst_face_rect = detect_face_lst(img)

            # append img name like '2002/08/11/big/img_344'
            lst_write2_fddb_ret.append(img_name)

            face_num = len(lst_face_rect)
            # append face num, note if no face 0 should be append
            lst_write2_fddb_ret.append(str(face_num))

            if face_num > 0:
                # append each face rectangle x y w h score
                for face_rect in lst_face_rect:
                    # append face rectangle x, y, w, h score
                    # note: opencv hava no confidence so use 1 here
                    s_rect = " ".join(str(item) for item in face_rect) + " 1"
                    lst_write2_fddb_ret.append(s_rect)

    except Exception as e:
        print 'error %s , can not generate complete fddb evaluate file' % e
        return -1

    # Write all the result to txt for FDDB evaluation
    write_lines_to_txt(lst_write2_fddb_ret)