AHRC RTI grant

[Earl G.P.]

Hi all,

 

I just gained some funding from the UK Arts and Humanities Research Council (AHRC) that will pay for development of an open source web viewer for RTI, and also a viewer for iOS6 that will be released to the community. The grant is small and won’t start until 1 June 2013 but I wanted to let you all know. I would really welcome input to the project so that we can all build on the RTI community. You can learn more here:

 

http://acrg.soton.ac.uk/projects/ahrc-rti-project/

 

The discussions about the project will have publically via the CHI forum linked from that page and we really welcome input.

 

All best wishes,

 

Graeme

 

---------------------------------------------------------------------------------------------

Dr Graeme Earl FSA

Web: http://about.me/graeme.earl

Twitter: http://twitter.com/graemeearl

---------------------------------------------------------------------------------------------

 

[Clifford Lyon]

Hi Graeme,

I wrote the Java version of the online PTM viewer, and I believe given the new HTML5 extensions, it should be possible to do everything the Java Applet version did natively in the browser--or at least some browsers.  I did start working on it but have not had time to pursue it.  A key item is the JPEG decompression.

I don't understand exactly the difference between PTMs and RTI, but I would be happy to help how I can and you are welcome to the Java code here: https://code.google.com/p/ptmviewer/

See also: http://materialobjects.com/ptm/

Cheers,

Cliff

--
Group website is: https://sites.google.com/site/buildrtis/
---
You received this message because you are subscribed to the Google Groups "RTI_help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rti_help+u...@googlegroups.com.
To post to this group, send email to rti_...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Earl G.P.]

Hi cliff

That is fabulous! Thanks so much. We used your viewer a great deal in the past. I agree re: html5. What I hope is that we can create an html5 based viewer plugin that will sit within the IIP Image moo viewer as a means to deal with very large datasets and also to overlay other data eg standard multispectral.

Would you mind if I copy this thread to the CHI forum so that I have a record there? We are archiving discussions about the project there.

All the best and I will definitely be asking for your help!

Graeme

On 25 Feb 2013, at 19:42, "Clifford Lyon" <cliffo...@gmail.com<mailto:cliffo...@gmail.com>> wrote:

Hi Graeme,

I wrote the Java version of the online PTM viewer, and I believe given the new HTML5 extensions, it should be possible to do everything the Java Applet version did natively in the browser--or at least some browsers. I did start working on it but have not had time to pursue it. A key item is the JPEG decompression.

I don't understand exactly the difference between PTMs and RTI, but I would be happy to help how I can and you are welcome to the Java code here: https://code.google.com/p/ptmviewer/

See also: http://materialobjects.com/ptm/

Cheers,

Cliff

On Fri, Feb 22, 2013 at 10:43 AM, Earl G.P. <graem...@soton.ac.uk<mailto:graem...@soton.ac.uk>> wrote:
Hi all,

I just gained some funding from the UK Arts and Humanities Research Council (AHRC) that will pay for development of an open source web viewer for RTI, and also a viewer for iOS6 that will be released to the community. The grant is small and won’t start until 1 June 2013 but I wanted to let you all know. I would really welcome input to the project so that we can all build on the RTI community. You can learn more here:

http://acrg.soton.ac.uk/projects/ahrc-rti-project/

The discussions about the project will have publically via the CHI forum linked from that page and we really welcome input.

All best wishes,

Graeme

---------------------------------------------------------------------------------------------
Dr Graeme Earl FSA
Web: http://about.me/graeme.earl
Twitter: http://twitter.com/graemeearl
---------------------------------------------------------------------------------------------


--
Group website is: https://sites.google.com/site/buildrtis/
---
You received this message because you are subscribed to the Google Groups "RTI_help" group.

To unsubscribe from this group and stop receiving emails from it, send an email to rti_help+u...@googlegroups.com<mailto:rti_help%2Bunsu...@googlegroups.com>.
To post to this group, send email to rti_...@googlegroups.com<mailto:rti_...@googlegroups.com>.

For more options, visit https://groups.google.com/groups/opt_out.

--
Group website is: https://sites.google.com/site/buildrtis/
---
You received this message because you are subscribed to the Google Groups "RTI_help" group.

To unsubscribe from this group and stop receiving emails from it, send an email to rti_help+u...@googlegroups.com<mailto:rti_help+u...@googlegroups.com>.
To post to this group, send email to rti_...@googlegroups.com<mailto:rti_...@googlegroups.com>.

[Clifford Lyon]

Sure, feel free to copy the thread.

Does the RTI data structure uses JPEG compression for the "planes", do you know?  

To unsubscribe from this group and stop receiving emails from it, send an email to rti_help+u...@googlegroups.com.
To post to this group, send email to rti_...@googlegroups.com.

[Lindsay MacDonald]

Good work, Cliff. The original PTM file format, which I'm
very familiar with, has 9 bytes per pixel for the LRGB format.
Three bytes represent the normalised colour image in red, green
and blue channels (same colour space as the original image),
and these could be de/compressed with JPEG (or any other
method) without compromising the result in the viewer.

But the 6 bytes per pixel for the luminance component
represent the coefficients of the biquadratic function that
models pixel intensity as a function of hemispherical light
direction. I think that if you tried to JPEG these you would
completely wreck the image. Similarly for bigger numbers
of coefficients for spherical harmonics in the RTI format.
It would be easy to test this by treating the coefficient
'planes' as independent monochrome images.

Tom Malzbender probably knows a lot more about this
subject (smile).

Lindsay MacDonald
Photogrammetry, 3D Imaging and Metrology Research Centre
UCL

--

[Clifford Lyon]

Right - iirc, there was another file format that used JPEG compression for the RGB channels.  The coefficients were stored separately.  There was a header that indicated the storage method.  Overall this reduced the size of the PTM file substantially, and did a pretty good job preserving the quality, though it is lossy.

I just mention it because when I saw the HTML5 features a couple years back, I started working on a Javascript/HTML5 PTM viewer, and that is where I got hung up.  

A big challenge with the old viewer was memory management (don't believe what Java tells you about garbage collection :) and loading the image as fast as possible.  So the smaller size helped.  It could be that now, years later, these won't be such big hurdles, as we're working with faster chips, more RAM, and better browsers.

[Clifford Lyon]

Nope, I remembered incorrectly.  The coefficients are compressed.  See 2.9.8 at:

http://www.hpl.hp.com/research/ptm/downloads/PtmFormat12.pdf

[Lindsay MacDonald]

Thanks Cliff. Well that's a surprise. I guess that the compressibility
depends on the degree of correlation between adjacent pixels in
the coefficient planes. Has anyone done any studies of losses in
spatial image quality when de/compressing PTM files in this way?

Lindsay