So this project is something I have hinted at in the main forum and feel it more belongs here, this thread is to monitor my progress and maybe get some help along the way.
The ultimate goal will be a JARVIS like system that Iron man has. Its main area to be used is costume making (there is a lot of makers out there), the system will take voice commands and turn that into an action on the suit itself, for example open faceplate will open the well faceplate.
The system will also run an augmented reality side, this will take 2 live feeds from HD webcams and overlay information on them, so far I have face detection working, I plan on adding person detection and maybe some other nice little tricks like calling people via voice, loudspeaker etc the augmented reality feed will be send to two tiny HD screens to be mounted in the eyes of the helmet. just in case you were wondering
In the end you user will hopefully be able to include this in their builds and use the suit more authentically and customize the commands according to how much they animate the suit.
I have a long road to go down so all feedback positive and negative is welcome, i also have a lot to learn in ways of speech recognition and image related topics (finding people).
now attached to this is the first ROUGH program, it takes a webcam feed (i have set it to camera 1 as that's what my laptop is, please change the camerasel value if you are playing with it), it also has the command open faceplate and close faceplate. It gives a status of the faceplate on screen but this will not be needed.
I had issues with the speed of voice recognition so I changed the code and it now works much better.
For the attached to work you will need to install and reference the Aforge.Net libaray and the Accord.Net libaray
My CodeBank Submissions
Listbox with transparency and picture support - Click Here
Reply With Quote