Amazing. I’ve installed the Photosynth preview on my own machine (sadly it seems to work in IE only on a PC—not surprising, but a little disappointing), and I am zooming around in the Piazza San Marco courtesy of photos shot by a Photosynth Program Manager. The experience is incredible, and totally unique.
There are questions that arise: Is participation something that is voluntary, or is it something more ubiquitous and automatic that will just happen when you upload pictures to the web? (In the case of the preview that I’m running, we can assume it was a Microsoft sponsored trip. But the question is pertinent for future plans.) What are the mechanisms in place to provide privacy? What are the mechanisms to allow for editorializing; for instance, what if I wanted to see only shots taken at night? The images I’m looking at of Saint Mark’s Plaza were all shot by the same person on what looks like the same day with the same camera. How will this work with a different set of images taken with different hands, shutter speeds, attention to details like focus, lighting, foregrounding, etc.? And a larger, geographical and geopolitical question: how were these sites chosen? Will we (the public) be able to contribute models as well as photos so that I can make my city block a photo-navigable space? Or, more importantly, someone in São Paulo can make a map of their city block?
But aside from the questions, this is the most exciting way to view photos from the ‘net that I have ever seen.
if:book
A Project of the Institute for the Future of the Book
Thanks for noticing our app. 🙂
I’m the one who took the pictures – and no, unfortunately I didn’t get paid for my trip – it was just an awesome vacation.
To answer a couple of your comments… First, we have a Firefox plugin! It should work fine for you. (no we don’t have a Mac version yet – it’s just a tech preview for now.)
My Venice photos were actually shot on different days – the weather was just so nice you can’t tell. We have to use sets like this for now, just because of image copyright issues. But, if you look at the collection we have in Korea, that one includes user submitted photos – and in fact that makes it more interesting. The source of the images makes no difference – it’s the overlap of the scene that lets us generate the model.
We hope to release something more interactive at some point, but we’re a small team within Microsoft and don’t always get what we want.
cheers!
jonathan
awesome, on several different levels. kudos!
-bowerbird
Photosynth’s definition, “the newest – and, we hope, most exciting – way to view photos on a computer” is a humble one. Beyond the sheer pleasure of the virtual walk inside a photo, is the fact that this is done through collective eyes. This multi-resolution experience, enhanced by the way the visual algorithms register the images together, is visualization as social environment. The content inside the images grows in complexity as people use them. It becomes exceptionally rich collective memory, thanks to metadata that people enter. As Blaise Aguera y Arcas concludes in his TED presentation, the potential to create these amazing virtual models of places is “the classic networked effect.” Some of the questions that Jesse presents are quite pertinent, but Photosynth’s possibilities are remarkable.
I also loved Seadragon’s ability to manage enormous amounts of visual data as well as its representation of non-image data, such as books and periodicals. The fact that what one has on the screen is real text, not image, delivers the “real thing.” And, the capacity to embed vast amounts of images, or text, seems to be a great alternative to the current limits of the screen.