helloFace is a face-detection tool which aims on simple setting and easy using. (source code)

• using OSC to send out the result of face-detection or to response query
• also receive OSC message for dynamic settings or query
• providing GUI for simple setting and easy using

1. install GStreamer framework on your platform (reference)
   • make sure that you can run video-related examples in Processing 2.x (if you can, that means you install GStreamer framework already.)
2. install OpenCV 2 on your platform (reference)
   • windows (中文參考)
   • Mac (中文參考)
   • Linux (中文參考)
3. download, unzip adn run helloFace app from links below:
   • Linux (32bit)
   • Linux (64bit)
   • Mac (64bit)
   • Windows (32bit)
   • Windows (64bit)

####about helloFace main app

• Press key [s/S] to switch on/off SettingPanel
  • to select/set camera paramerters
  • to set OSC-related paramters
  • to set current settings as default settings for next usage
  • to delete current default setting files
• Press key [i/I] to switch on/off InfoPanel
• Press key [v/V] to switch on/off visual hints of detection
  • always send out OSC messages of detecting result, whether the visual hints is on or off

####about helloFace client app you can use any language or tools that support OSC protocol to build a OSC connection with helloFace app.
, and then you can get the OSC messages about results of detecting faces or your query from helloFace.
in the future, you also can send OSC message to helloFace for advanced using.

####download helloFace client Examples now some exanples are provided, you can download for testing and get more understanding about helloFace

• Processing examples

####helloFace send [ OUT ]

• about responding camera's resolution

  • address pattern：/returnCamRes
  • type tag：ii
    • i: first one stands for camera's width
    • i: second one stands for camera's height
  • Example：/returnCamRes 640 480

• about responding camera's FPS

  • address pattern：/returnCamFps
  • type tag：i
    • i: stands for camera's FPS (frame per second)
  • Example：/returnCamFps 30

• about face-detection

  • address pattern : /faceDetect
  • type tag: is
    • i: total count of detected faces
    • s: list of detected faces' ranges (rectangle range). Format is x=0.4640625,y=0.47083333,w=0.1828125,h=0.24375|x=0.4453125,y=0.35416666,w=0.2484375,h=0.33125|...（and so on），every face range is separated by | , every face range (rectangle range) is consist of 4 numbers (every value is normalized，value is between0 and 1): x stands for the x position of top-left corner of rectangle, y stands for the y position of top-left corner of rectangle, w stands for the width of rectangle, h stands for the height of rectangle
  • Example：/faceDetect 2 x=0.525,y=0.52708334,w=0.09375,h=0.125|x=0.4421875,y=0.36875,w=0.2484375,h=0.33125

####helloFace receive [ IN ]

• about receiving query of camera's resolution

  • address pattern：/queryCamRes
  • type tag：none

• about receiving query of camera's FPS

  • address pattern：/queryCamFps
  • type tag：none

helloFace uses GStreamer framework to capture live camera video, and then uses OpenCV 2 to do face-detection.

• Processing
• javacvPro library for getting the power of OpenCV 2
• oscP5 library
• controlP5 library