PrimeSensor Modules for OpenNI (Unstable branch)
------------------------------------------------

Kinect Mod:
-----------

***** Important notice: *****
You must use this kinect mode version with the unstable OpenNI release which is available at:
http://www.openni.org/downloadfiles/openni-binaries/20-latest-unstable


Changes:
v0.4:
	1) Merged code with the new unstable sensor driver from PrimeSense that adds MacOSX and multiple sensor support.
	2) Added precompiled binaries for Linux & MacOSX.

v0.3:
	1) Added support for high-res 1280x1024 IR and Image (at ~10 FPS)
	   Checkout the file OPENNI\Data\SamplesConfig.xml for usage example.
	   (Copy it to your OpenNI\Data dir and use NiViewer to test the different resolutions)

v0.2:
	1) VS2010 redist will now be installed automatically.
	2) The driver will now also be installed automatically on both x86 and x64 machines.
	3) Turned off the IR projector anti-cover thingy. It is off on the Xbox360 so it must be safe and it was very annoying...

v0.1:
	1) Added the kinect USB VID/PID to the PSDrv3 driver and to the code.
	2) Changed the packet sequence field from 16-bit to 8-bit (don't know why it's different...)
	3) Changed the image format from YUV to 8-bit uncompressed bayer in the GlobalDefaults.ini file.
	4) Fixed the software depth-to-RGB registration. (Thanks ROS.org for that!)

Install notes:
	1) Install unstable OpenNI (http://www.openni.org/downloadfiles/openni-binaries/20-latest-unstable)
	   On Linux/Mac please do: ./sudo install.sh
	2) Install Sensor (this version...)
	   On Linux/Mac please do: ./sudo install.sh
	3) Install unstable NITE (http://www.openni.org/downloadfiles/openni-compliant-middleware-binaries/33-latest-unstable)
	   On Linux/Mac please do: ./sudo install.sh
	   Use this license when asked during the installation: 0KOIk2JeIBYClPWVnMoRKn5cdY4=
	4) Test #1: Run the NiViewer sample to make sure depth & image streams are working.
	5) Test #2: Run the OpenNI/NiUserTracker sample play with the skeleton. Don't forget to start with the calibration pose! (Explained in the PDFs)
	6) Test #3: Try the NITE/Sample-PointViewer sample for the hand tracking demo.

	For Mac/Linux, you will need to install the latest LibUSB.
	Linux:
		1) Run: sudo apt-get install libusb-1.0-0-dev
		2) Run: sudo apt-get install freeglut3-dev
	MaxOSX:
		1) Intall MacPorts from: http://distfiles.macports.org/MacPorts/MacPorts-1.9.2-10.6-SnowLeopard.dmg
		2) Run: sudo port install libtool
		3) Run: sudo port install libusb-devel +universal
		   Note: Do not forget the +universal, it's very important!!
		   If you're previously already installed libusb-devel then use "sudo port uninstall libusb-devel" and reinstall it again with the +universal flag.
	
	Note: All of the NITE samples are using 320x240 resolution. You need to change it to 640x480 in the XML files inside the Data directory.
	      I have prepared preconfigured XMLs in the "NITE\Data" dir. Just copy them to "c:\Program Files\Prime Sense\NITE\Data".

TIPs:
	1) If you want to see the IR stream, edit the "OpenNI\Data\SamplesConfig.xml" file and change all the "Image" words into "IR".
	2) If the driver is installed correctly you should see "Kinect Camera" and "Kinect Motor" under the "PrimeSensor" category in the device manager.
	   Please note that the audio device is not supported yet.
	
Todo:
	1) Add motor & led control.
	2) Add audio.

The source code for this version is available at:
https://github.com/avin2/SensorKinect	
	
-------------------------- ORIGINAL README BELOW --------------------------

Website: www.primesense.com
Forum: http://groups.google.com/group/openni-dev
Wiki: Coming soon!

Unstable branch new features:
* Audio streams are now supported.
* Sensor Server now supports multiple sensors. 
* Added MacOSX platform support.
* Improved README! (much more detailed instructions...)

Unstable branch fixes:
* Server: a potential dead lock was solved.
* GetStringProperty would always return XN_STATUS_ERROR.
* Bug fix: infinite loop in server when a client tries to disconnect and streams can't be closed (usb timeout for example).

Binaries are available at:
http://www.openni.org/downloadfiles/openni-compliant-hardware/31-latest-unstable

Please use this with the OpenNI unstable branch located at:
Source:
https://github.com/OpenNI/OpenNI/tree/unstable
Binaries:
http://www.openni.org/downloadfiles/openni-binaries/20-latest-unstable


Release Notes:
--------------
* At the moment, the default is to compile the code with SSE3 support (this is also true for the supplied binaries). 
  If you have a CPU without such support, please remove the sse compiler flags from the make files.
  (A good hint for this error is that you encounter an "illegal instructions" messages)
* By default, the linux kernel mounts unknown USB devices with write permissions to root only, and read-only permissions to other users.
  When not running as root, this prevents communicating with the device. To fix this issue:
  - Navigate to Platform/Linux-x86/Build
  - run 'make install-usb-rules'
  - if the device was connected, unplug and reconnect it.
  This part is done automatically by the install script, but you will need to do it manually if you choose not to use the automated install.
* The device driver contains a thread for reading from the USB device. This thread should have high priority in order to function correctly. Each process using the device driver should have the CAP_SYS_NICE capability in order to raise thread priority. Failing to have that capability, will cause a failure in setting priority, consequently causing loss of data.
* MacOSX: Only OSX 10.6 (Snow Leopard) with a 64-bit Intel based CPU is currently supported.
  Client/Server is not supported yet so do not attempt to access the sensor with more then one application.
  
  
Build Notes:
------------
Windows:
	Requirements:
		1) Microsoft Visual Studio 2008/2010
		   From: http://msdn.microsoft.com/en-us/vstudio/bb984878.aspx
		2) Python 2.6+/3.x
		   From: http://www.python.org/download/
		3) PyWin32
		   From: http://sourceforge.net/projects/pywin32/files/pywin32/
		   Please make sure you download the version that matches your exact python version.
		4) NSIS 2.46
		   From: http://sourceforge.net/projects/nsis/files/NSIS%202/2.46/
		5) NSIS-2.46-strlen_8192.zip patch
		   From: http://sourceforge.net/projects/nsis/files/NSIS%202/2.46/
		   This is a required patch to fix a critical bug in NSIS with long environment variables.
		   Simply unpack it into the directory where NSIS is installed and overwrite all files.
		6) OpenNI v1.x
		   From: http://www.openni.org/?q=node/2	

	Optional Requirements (To build the documentation):
		1) Doxygen
		   From: http://www.stack.nl/~dimitri/doxygen/download.html#latestsrc
		2) GraphViz
		   From: http://www.graphviz.org/Download_windows.php
		   
	Building Sensor:
		1) Go to the directory: "Platform\Win32\CreateRedist".
		   Run the script: "Redist.py".
		   This will compile and prepare the redist exe files that includes everything.
		2) Install the exe you've just made which is located in Platform\Win32\CreateRedist\Final\Output\Sensor-Win32-5.x.x.x.exe
		   The installer will also automatically register all the modules into OpenNI via the NiReg utility.
		   
		The visual studio solution is located in: Platform\Win32\Build\EngineWin32.sln.
		
		Important: Please note that even though the directory is called Win32, you can also use it to compile it for 64-bit targets (AMD64/x64).
   
Linux:
	Requirements:
		1) GCC 4.x
		   From: http://gcc.gnu.org/releases.html
		   Or via apt:
		   sudo apt-get install g++
		2) Python 2.6+/3.x
		   From: http://www.python.org/download/
		   Or via apt:
		   sudo apt-get install python
		3) OpenNI v1.x
		   From: http://www.openni.org/?q=node/2
		   
	Building Sensor:
		1) Go into the directory: "Platform/Linux-x86/CreateRedist".
		   Run the script: "./RedistMaker".
		   This will compile everything and create a redist package in the "Platform/Linux-x86/Redist" directory.
		   It will also create a distribution in the "Platform/Linux-x86/CreateRedist/Final" directory.
		2) Go into the directory: "Platform/Linux-x86/Redist".
		   Run the script: "sudo ./install.sh" (needs to run as root)

  		   The install script copies key files to the following location:
		       Libs into: /usr/lib
		       Bins into: /usr/bin
		       Config files into: /usr/etc/primesense
		       USB rules into: /etc/udev/rules.d 
		       Logs will be created in: /var/log/primesense
			
		To build the package manually, you can run "make" in the "Platform\Linux-x86\Build" directory.
		
		Important: Please note that even though the directory is called Linux-x86, you can also use it to compile it for 64-bit targets and pretty much any other linux based environment.	

MacOSX:
	Requirements:
		1) Xcode 3.2.5
		   From: http://developer.apple.com/devcenter/mac/index.action
		   https://developer.apple.com/ios/download.action?path=/ios/ios_sdk_4.2__final/xcode_3.2.5_and_ios_sdk_4.2_final.dmg
		   Please note that you need to register as a mac developer (It's free!).
		2) OpenNI v1.x
		   From: http://www.openni.org/?q=node/2	   
	   
	Building Sensor:
		1) Go into the directory: "Platform/Linux-x86/CreateRedist".
		   Run the script: "./RedistMaker".
		   This will compile everything and create a redist package in the "Platform/Linux-x86/Redist" directory.
		   It will also create a distribution in the "Platform/Linux-x86/CreateRedist/Final" directory.
		2) Go into the directory: "Platform/Linux-x86/Redist".
		   Run the script: "sudo ./install.sh" (needs to run as root)

  		   The install script copies key files to the following location:
		       Libs into: /usr/lib
		       Bins into: /usr/bin
		       Config files into: /usr/etc/primesense
		       USB rules into: /etc/udev/rules.d 
		       Logs will be created in: /var/log/primesense
			
		To build the package manually, you can run "make" in the "Platform\Linux-x86\Build" directory.