candybot_video_recognition

Module for CandyBot to help recognise people and objects via web-cam

Automatic installation:

run install.sh: ./install.sh

if OpenCV already installed then comment OpenCV installation section in install.sh

Manual installation:

Libraries list and intallation instruction:

I. Python libraries:

1. rospkg, catkin_pkg - ros libraries for Python 3.5:

	sudo pip3 install rospkg, catkin_pkg
	
2. pocketsphinx - speech recognition library:

	sudo pip3 install pocketsphinx
	
3. pyaudio - audio record library, depends on avbin package https://github.com/downloads/AVbin/AVbin/install-avbin-linux-x86-64-v10:
	
	sudo pip3 install pyaudio
	
4. wikipedia - Python API Wiki library:
	
	sudo pip3 install wikipedia

II. Non-Python libraries:

1. ROS Kinetic (official recommended way):
	
	1.1. follow link http://wiki.ros.org/Installation
	
	1.2. choose your OS
	
	1.3. follow instructions
	
2. OpenCV:
	
	You can use two links:
	
		- instruction from official site http://docs.opencv.org/3.1.0/d7/d9f/tutorial_linux_install.html
		
		- short instruction from askubuntu http://askubuntu.com/questions/783956/how-to-install-opencv-3-1-for-python-3-5-on-ubuntu-16-04-lts
		
		WARNING! : you can use both links following your intuition... and knowledge, of course 
		WARNING! : if Python 3.5 already installed in your system pass Python-installation steps
		ACHTUNG!! : DO NOT USE OPENCV INSTALLATION INSTRUCTION FOR PYTHON VERSION LESS THAN 3.5
	
3. RHVoice - speech synthesis library:

	https://launchpad.net/~linvinus/+archive/ubuntu/rhvoice

---