| ----1 This talk is entitled Worlds within Worlds: it's a great title -- it raises so many questions -- but I'll mainly be talking about panoramas I'm Graham Mills, a university biologist from Liverpool, UK Bad news: I'm not an expert at panoramas :( Good news: You don't have to be :) First up, what do *I* mean by a panorama in this context? The first slide shows an exhibit at the Tech Museum in SL(Second Life) that focuses on how artists have represented 3D images in 2D space It's an excellent museum if you haven't been there To a photographer, a panorama might mean just a wide angle view of a landscape SL on the other hand is intrinsically panoramic -- you can generate images to encompass the full 360 degrees around you -- so I'm taking that (or something close it) as my starting point. The ability to relate to this panoramic "feel" helps give teacher and student a sense of shared presence. So I'm taking panorama in SL to mean an image that represents a <--wide--> field of view AND that is projected in such a way that we have an illusion of depth and hence that vital sense of presence. Personally, in RL (real life) it can seem a bit weird to be taking pictures of the ground by your feet when other people are about :) ----2 The panorama is typically displayed as a texture (SL image) on the inner surface of a box shape, on a sphere or a cylinder The width or height of the texture means that it is normally a composite of multiple images stitched together This stitching might mean using something like the Photoshop Photomerge function or it might use specialist tools The NMC recently negotiated the FREE use of Stephane Zugzwang's Virtual Reality Room (VRR) on educational sims and that's an example of the box shape Here you can see my avatar, suitably dressed, apparently in the Chateau de Villandry in France Getting the illusion of 3D inside a box means that you have to use a projection, in the VRR's case it is the cubic projection used by QuickTime VR So you need a second piece of software, in my case on Windows a freebie called Pano2QTVR, to produce the six sides of the cube For the VRR you can buy commercial photosets (like the Villandry set) and, if you want to mount your own, the HUD ( (HUD = SL Heads-Up Display) to do this as well The NMC kit (including a range of VRR sizes) is available on their Learning sim and comes with instructions on how to make your own Eloise Pasteur has scripted a small (freebie) prim to automate the process a little further still (some examples at my avatar's feet) You just drag your textures onto the appropriate surface of the prim, touch it and the image is projected onto the walls of the VRR You do, however, need to edit the script do that it uses the appropriate communication channel -- this restricts who can use your VRR This makes it as easy as using the commercial photosets; you can save the prim, reuse it, put it in a sequence, swap it with other educators, etc ----3 Too much info already? Want to digest at your leisure? When it comes to understanding a little more about panoramas, I recommend checking out the PanoHUD toolkit by SL avatar Richard Blackhawk, part of an exhibit called Reprojecting San Jose, formerly in The Tech Museum but now on Carrick. Richard has a little booth where you can get some background on the technique as well as the toolkit itself The toolkit gives concise guidance on making both RL (real life ) and SL panoramas As its name suggests, the PanoHUD automates image capture for SL panoramas. Highly recommended The main image in the booth shows the RL panoramic image and at the bottom right is the huge Circle of Palms as rezzed on Carrick ----4 The VRR comes in a range of sizes but one issue with it perhaps is the hermetically sealed nature of the experience: it takes over your entire field of view in SL If you want students to see a little of multiple environments *simultaneously* then you could use a phantom cylinder textured internally with a PanoHUD image The image is on the inner surface and outside is set to transparent so as your avatar walks round the view changes. If you go inside, you can swivel to see the full 360 This is an image of my old parcel in EduNation seen, erm, on EduNation. Given that I'm not there any more and SL changes so quickly, it is a record as much as a personal memento You could, however, use something like this for briefing students ----5 Spheres also work well, especially if you process out the pinched effect at the top using a Photoshop plugin developed by Richard Rosenman The plugin works just as well in the freeware IrfanView This view is of part of Hadrian's Wall of which more anon ----6 There is a commercial version called SceneGlobe that comes with nice freebie image sets: you touch the sphere to move between images The avatar sits in the centre and the SceneGlobe optionally rotates -- some people find this makes them a little queasy but you can turn it off These are low prim and you can as a consequence have a large number so that the spheres might comprise, say, waypoints on a tour ----7 The neogeographers from the UCL CASA group in London have exploited their own version of the sphere to produce larger urban walk-in "bubbles" that can be used to map data to specific visual locations. You can see these on a sky platform on Second Nature ----8 Motion can also be obtained by putting video on the walls and this has been done in a huge example on Navaria by Quality Innovations The end-result is currently, I think, in the Linden Lab Showcase and combines video with music to great effect ----9 Sometimes unexpected views work surprisingly well This is Google Earth mapped onto the VRR by Corro Moseley and colleagues over Commonwealth Island It is, of course, inside out but we know the subject so well that it makes perfect sense They have used it to display news feeds from different geolocations ----10 Of course, it is one further step to displaying live feeds as imaginary images, exploiting our inherent ability to recognise features and patterns to help make broad sense of otherwise complex and fluctuating data While I don't know of any examples using the VRR, this 2D panorama from Bashiba.com is a visualization of the stock market as a sea and landscape streamed into SL under the auspices of Melanie Swan (SL: Xantha Oe) We can probably guess what the video stream might look like at the moment weather-wise ;) ----11 This is fun but what educational need do these panoramas/projections meet? Well, firstly, they are easy and inexpensive to produce so teachers with limited experience who are basic-level photographers (like me!) can contribute without having to learn to build in SL If we extend that idea to students, we have user-generated content that might record, say, a field visit With a little more experience, as shown here, you can have a build (a cross-section of the Roman Hadrian's Wall in the north of England) and a VRR panorama that complement one another (photos by Dr Alan Greaves, reconstruction by Eloise Pasteur) Thus students can see the wall in its original state as well as the wall as it is now, and appreciate its present-day topographical context Some people are concerned that panoramas deskill: this is an example where they complement conventional approaches ----12 This image is a view of our campus at Liverpool I guess it might be used as part of an orientation for students prior to arrival ----13 However, it is also important to appreciate the limits of panoramic projections Sometimes there is inevitable distortion, especially within confined spaces as with this rendering of the Mercury capsule at the International Space Museum It is still, however, a powerful image Personally, I wouldn't go to the local store in this, let alone into space :) The nature of the geometry in panoramas often makes it difficult to graft on prim-based extensions Finally, the very richness of the environment may enhance engagement but make it harder to identify the primary features of interest Of course, we could use post-processing to produce variants with labels to primary features or reduced levels of non-specific detail ----14 To finish off, I just want to mention a couple of other ways of representing scenes in SL Firstly, here is a sculpty terrain prim on Max Chatnoir's Genome You can get this done as a service but there is also a freebie kit ----15 The final slide shows a couple of new ways people are exploring this 3D in 2D conundrum. Firstly, on the left we have Microsoft Photosynth (this example, FloraPrims, is by Vlad Bjornson and comprises 344 still images!) As we can see, Photosynth can provide views of SL builds *outside* the SL client in an aesthetically interesting and flexible alternative, say, to QTVR or machinima, ie video Photo Tourism, on the right, is a related Microsoft project that aggregates images to provide paths through the landscape of images To end, I just want to summarise by saying that, used with care, panoramas have the ability to bring new images and hence experiences, from the both real and virtual world, to students as both consumers and producers And they're fun! Thanks for reading these lines (courtesy of Dudeney Ge's speakEasy HUD). Any questions? Remember, I'm not an expert :) |