becta logo
[oats-sig] open gazer problems

stephen emslie stephenemslie at gmail.com
Sun Nov 11 22:17:01 GMT 2007

Article: [oats-sig] open gazer problems

Hi Simon. I rather optimistically took a stab at porting/rewriting
their visible-spectrum algorithm in python+numpy a while back, as I
dont have a matlab license (and it seems a bit strange that an open
source project requires an expensive license to run). I really like
their ideas, but its a shame about it being matlab only. I got
distracted when I discovered opengazer, but perhaps it would be worth
picking it up again, though it would possibly be just for the sake of
learning more.

Compiling it with Octave was another idea, but when I tried it didn't
quite support enough of matlab's functionality. Perhaps thats changed
now.

Stephen Emslie

On Nov 9, 2007 6:00 PM, Simon Judge <simon.judge at nhs.net> wrote:
> Someone just reminded me of this too, if anyone wants to get their hands
> really dirty!!!
>
> http://thirtysixthspan.com/openEyes
>
>
>
>
> Cheers.
>
> Simon
>
>
> -----Original Message-----
> From: oats-sig-bounces at lists.becta.org.uk
> [mailto:oats-sig-bounces at lists.becta.org.uk] On Behalf Of Steve Lee
> Sent: Thursday, November 08, 2007 10:47 AM
> To: OATs Project Special Interest Group
> Subject: Re: [oats-sig] open gazer problems
>
> Perhaps this demo of MyTobii and the SmartBox's Grid2 will inspire you.
>
> http://uk.youtube.com/user/mobiletobii
>
> One thing that really struck me is how far we have to go with optimising
> this mode of input. It takes conciderable time to createa very short phrase.
>
> http://www.smartboxat.com/mytobii.html
>
> Steve
>
> On 07/11/2007, stephen emslie <stephenemslie at gmail.com> wrote:
> > Great that you got it working! It took me quite a while to get the
> > feel of compiling against vxl too.
> >
> > I would love to hear any feedback on what sort of accuracy you are
> > able to achieve and what webcam you are using.
> >
> > Stephen Emslie
> >
> > On Nov 7, 2007 3:11 PM, Chris Feser <chrimp at gmx.de> wrote:
> > >
> > > Thx Steve!
> > > Now I understand the connection from vxl to opengazer.
> > > I fixed it quick and dirty and copied the missing files to the
> > > expected directory. I have no idea what I did wrong the whole time, but
> now it works.
> > >
> > > thank you all very much for your help!
> > >
> > > I wrote the developer of opengazer an email if there is a support
> > > forum or a mailing list, but I have no answer yet.
> > > Does anybody of you have an idea or is this forum the best place for
> > > opengazer problems?
> > >
> > > I'll play around with opengazer and try to extend it or even will
> > > try, to port it to Windows.
> > >
> > > cheers and have a nice day!
> > >
> > > chris
> > >
> > >
> > >
> > >
> > > Steve Lee-3 wrote:
> > > >
> > >
> > > > Hi Chris, yes, and to give a little more detail to my previous
> reply....
> > > >
> > > > taking theLeastSquares.h:2:28  error as an example Look in the
> > > > file LeastSquares.h line 28 it will have a line something like
> > > > #include <vnl.vnl_vector.h> or #include "vnl.vnl_vector.h"
> > > >
> > > > I'm guess  from the error it has <vnl/vnl_vector.h>.
> > > >
> > > > The critical thing is if it is prefixed 'vnl/'. if so then the
> > > > compiler will search for a file 'vnl/vnl_vector.h' under each of
> > > > the paths specified in you makefile with INCLUDES. In other words
> > > > vnl_vector.h is sought in a folder vnl under each given path.
> > > > (if the #include "file" form is used life is a little more complex
> > > > but we can probably ignore that)
> > > >
> > > > If your vnl_vector.h is not in such a folder then you can either
> > > > copy all of them for a quick fix or change the makefile to change
> > > > where the compiler is told to look.
> > > >
> > > > I'm afraid I'm tied up at the mo and can't check the real files
> > > > but perhaps stephen can shed a little more light on the matter?
> > > > * stephen where are your vnl headers?
> > > > I have Gutsy so will try, but it probably won't be till next week.
> > > >
> > > > Steve
> > > >
> > > > On 06/11/2007, Chris Feser <chrimp at gmx.de> wrote:
> > > >>
> > > >> Thx.
> > > >> I think I have a problem wih the header files.
> > > >> Should they be there, where I compiled VXL into?
> > > >> BUILD_SHARED_LIBS are ON and I changes the  CMAKE_INSTALL_PREFIX
> > > >> to /usr/bin because I compiled VLX now into this directory.
> > > >>
> > > >> Could yo please tell me, where are the header files should be?
> > > >>
> > > >> I use ubuntu 7.10 Gutsy
> > > >>
> > > >> cheers
> > > >> chris
> > > >>
> > > >>
> > > >> Stephen Emslie wrote:
> > > >> >
> > > >> > Now that I'm home perhaps I can compare my own Makefile.
> > > >> >
> > > >> >> but my compiled vxl is there:
> > > >> >>  /home/chris/Desktop/software/bin
> > > >> > ...
> > > >> >> My Makefile looks like this:
> > > >> >> # required libraries: vxl, opencv, boost, gtkmm
> > > >> >>
> > > >> >> VXLDIR = /home/chris/Desktop/software/bin
> > > >> >
> > > >> > It could be that the header files built when installing vxl are
> > > >> > not in /home/chris/Desktop/software/bin? When you configured
> > > >> > vxl with cmake, what was "CMAKE_INSTALL_PREFIX" set to (mine is
> > > >> > /usr - same as VXLDIR)? Obviously "BUILD_SHARED_LIBS" needs to
> > > >> > be on for this to work.
> > > >> >
> > > >> > The only thing I really had to change in the Makefile was to
> > > >> > set the VXLDIR to the same as the install prefix.
> > > >> >
> > > >> > Best of luck
> > > >> >
> > > >> > Stephen Emslie
> > > >> >
> > > >> >
> > > >> >
> > > >>
> > > >> --
> > > >> View this message in context:
> > > >> http://www.nabble.com/Re%3A-open-gazer-tf4558293.html#a13615292
> > > >> Sent from the OATS Project Interest Group mailing list archive at
> > > >> Nabble.com.
> > > >>
> > > >>
> > > >>
> > > >
> > > >
> > > > --
> > > > Steve Lee
> > > > --
> > > > Jambu - Alternative Access to Computers www.fullmeasure.co.uk
> > > >
> > > >
> > > >
> > >
> > > --
> > > View this message in context:
> http://www.nabble.com/Re%3A-open-gazer-tf4558293.html#a13628973
> > >
> > > Sent from the OATS Project Interest Group mailing list archive at
> Nabble.com.
> > >
> > >
> > >
> >
> >
>
>
> --
> Steve Lee
> --
> Jambu - Alternative Access to Computers
> www.fullmeasure.co.uk
>
>
>
> **********************************************************************
> This message  may  contain  confidential  and  privileged information.
> If you are not  the intended  recipient please  accept our  apologies.
> Please do not disclose, copy or distribute  information in this e-mail
> or take any  action in reliance on its  contents: to do so is strictly
> prohibited and may be unlawful. Please inform us that this message has
> gone  astray  before  deleting it.  Thank  you for  your co-operation.
>
> NHSmail is used daily by over 100,000 staff in the NHS. Over a million
> messages  are sent every day by the system.  To find  out why more and
> more NHS personnel are  switching to  this NHS  Connecting  for Health
> system please visit www.connectingforhealth.nhs.uk/nhsmail
> **********************************************************************
>
>
>


  Main Becta Site  | Return to top