• find face on picture
• blur founded box with face
• recursive search all valid file(.jpg|.png) and put it to image processing
• create shared lib in linux 31.08.19 update - lib placed in branch libfdetect
• save coords to json update 1.09.19 - create with boost::property_tree::ptree
• json multi write ready
• shared lib ready (branch libfdetect)
• save resize pictures ready
• create .dep package ready (libfdetect_dep)
• add tests
• add async work for face find and generate

1. .json file with coords has named as original picture - boost function json_write rewrite nodes
2. works on Ubuntu 18.04
3. compiler GCC 7.4.1
4. cmake version - 3.10
5. boost version - 1.65.1
6. OpenCV version 3.4
7. STL C++ - 11

result:

cfsawarenessday_1024px.jpg.json:

{
    "0": [
        "366",
        "62",
        "472",
        "168"
    ],
    "1": [
        "47",
        "97",
        "140",
        "190"
    ],
    "2": [
        "182",
        "148",
        "308",
        "274"
    ],
    "3": [
        "667",
        "155",
        "789",
        "277"
    ],
    "4": [
        "368",
        "240",
        "526",
        "398"
    ],
    "5": [
        "694",
        "420",
        "827",
        "553"
    ],
    "6": [
        "79",
        "421",
        "167",
        "509"
    ]
}