The intention of this project is to interact and stream frames from video devices and do facial analysis on them.
A C++11 compliant compiler (GCC 4.8+, Clang 3.3+ etc.), CMake 2.8.8+, Qt 5.2+ and OpenCV 2.4+.
To compile the source code and link the binaries do the following:
- Extract source and go into the diretory.
- mkdir build
- cd build
- cmake ..
- make
This produces two binaries in the bin folder: snapshot and faces
snapshot is used to snap an image from a video device and save it to disk.
Usage: ./bin/snapshot (<options>) <output filename>
The format of the snapshot is dictated by the extension of <output filename>.
Supported extensions: jpg, jpeg, and png.
Options:
--help | -h Shows this message.
--list | -l List all available video devices on the system.
--device | -d <str> The device to take a snapshot from.
--quality | -q <n> Quality of the snapshot in percentage (1-100).
--scale | -s <str> Scales to the defined size given 'WidthxHeight'
in pixels. Or scale keeping the size-ratio by
'Xw' or 'Xh', where 'X' is the size to scale to
in width or height respectively.
This utility is used to analyse for faces on one or several images.
Detects faces and eyes in an image.
Usage: ./bin/faces (<options>) <face file> <eyes file> <image file>
<face file> and <eyes file> must to be cascade files in XML that
are either LBP or HAAR types.
Options:
--help | -h Shows this message.
--overlay | -o <file> Write image with overlays to file.
--no-faces | -nf Don't draw faces to overlay.
--no-eyes | -ne Don't draw eyes to overlay.
--out-data | -od <file> Write faces to an XML file.