Chris_Malcolm posted a photo:

The sun is setting over the west side of George Square, which has in effect become the campus of what used to be called the Faculty of Arts of Edinburgh University, and now is called the College of Humanities & Social Science. But not quite, because reflected in the windows of the Appleton Tower is the Informatics Forum. That houses the research side of the School of Informatics, a coalition of what used to be the separate disciplines of Computer Science, Cognitive Science, Computational Linguistics and Artificial Intelligence.

That is part of the College of Science & Engineering, most of which is located at King's Buildings, the science campus. These two Colleges are the modern form of C P Snow's famously separate and warring Two Cultures.

But knowledge is hard to categorise in a simple hierarchical tree structure. Like family relationships it forms multiple alliances and friendships across taxonomic boundaries. The late Professor McGonigle, whose Laboratory of Cognitive Neuroscience was once housed at the top of the Appleton Tower, used to refer to the Procrustean bricks and mortar taxonomy of science beloved of university administrators and literally fossilised in the stonework of university buildings, and which interdisciplinary research always confused. His late Laboratory of Cognitive Neuroscience collaborated across the former Faculties of Arts and Science with the late Department of Artificial Intelligence.

Now the Informatics Forum is part of the College of Science & Engineering, and the Dugald Stewart Building is part of the College of Humanities & Social Science. Despite belonging to the opposite camps of C P Snow's Two Cultures there is interdisciplinary collaboration between these two buildings. It's therefore quite appropriate that the Informatics Forum and the Dugald Stewart Building are adjacent, being part of the same Potter Row complex of buildings.

So the progressive dissolution of the boundaries between C P Snow's Two Cultures has now reached the stage of being incorporated in the bricks & mortar taxonomy of university building in the form of this new Potter Row development. You can look down on the whole interdisciplinary complex pf the Potter Row development from the site of Brendan McGonigle's old laboratory at the top of the Appleton Tower.

Original DSC01288RW_pt

Open photo page This is my first successful attempt at generating a tone-mapped image from an HDR image composed by blending different exposures. The two source images, which were hand-held shots with my Sony Ericsson k800i cameraphone (so JPEGs), are shown in the first comment below. The steps I used to generate the image were: * Use Hugin to align the two images based on automatically found SIFT landmarks * Click "Calculate Optimal Size" on the "Stitcher Page" * Make sure that these options are selected on the same page "Blended panorama (enfuse)", "Blended exposure layers", "Blended HDR panorama" and "Stacked HDR images" * Save the project in case anything goes wrong * Click "Stitch now!" * Look at the output called "-fused.tif" and find the region you want to crop to * Go back to the "Stitcher" page and enter the coordinates in the Crop options * Click "Stitch now!" again I found that if I didn't generate cropped output then the large black regions around the image caused the pfstools to fail in unexpected ways (e.g. all black output, an exception "terminate called after throwing an instance of 'std::out_of_range'") but YMMV, I suppose. Then make sure you have pfstmo, pfstools, pfsview and qtpfsgui installed (actually, you don't need the latter two, but they're fun to play with as well) and run: pfsinexr foo_hdr.exr | pfssize --maxx 1000 | pfstmo_fattal02 | pfsout test-output.png ... where foo_hdr.exr is the output from your second run of stitching. (You can view this on its own with pfsinexr foo_hdr.exr | pfsview.) You might want to adjust the --maxx parameter, since the fattal02 method behaves differently for different scales. Taken on 11th May, 2009 by oosp edinburgh,  hdr,  hugin,  appletontower ... Anyone can see this location

oosp posted a photo:

This is my first successful attempt at generating a tone-mapped image from an HDR image composed by blending different exposures. The two source images, which were hand-held shots with my Sony Ericsson k800i cameraphone (so JPEGs), are shown in the first comment below.

The steps I used to generate the image were:

* Use Hugin to align the two images based on automatically found SIFT landmarks
* Click "Calculate Optimal Size" on the "Stitcher Page"
* Make sure that these options are selected on the same page "Blended panorama (enfuse)", "Blended exposure layers", "Blended HDR panorama" and "Stacked HDR images"
* Save the project in case anything goes wrong
* Click "Stitch now!"
* Look at the output called "-fused.tif" and find the region you want to crop to
* Go back to the "Stitcher" page and enter the coordinates in the Crop options
* Click "Stitch now!" again

I found that if I didn't generate cropped output then the large black regions around the image caused the pfstools to fail in unexpected ways (e.g. all black output, an exception "terminate called after throwing an instance of 'std::out_of_range'") but YMMV, I suppose.

Then make sure you have pfstmo, pfstools, pfsview and qtpfsgui installed (actually, you don't need the latter two, but they're fun to play with as well) and run:

pfsinexr foo_hdr.exr | pfssize --maxx 1000 | pfstmo_fattal02 | pfsout test-output.png

... where foo_hdr.exr is the output from your second run of stitching. (You can view this on its own with pfsinexr foo_hdr.exr | pfsview.) You might want to adjust the --maxx parameter, since the fattal02 method behaves differently for different scales.

Antonis Lamnatos posted a photo:

Sign on the garage area of the Appleton Tower, Edinburgh.

I've always liked that sign, every since I saw it for the first time. I like the simple design, the clean font and the four, flat-coloured sections. The building might be voted amongst the ugliest in Britain, but that sign has caught my eye in a positive way ever since I first saw it. (Maybe I should be worrying that I talk like that for a sign?...)