I saw these the other day and wondered—could you make these programmatically with a drone? Have it lift off, tilt downwards, and stitch them together...
I get the sense that that's exactly how these were created. The transition from ground view to aerial view gives no hint of a full-sized aircraft's ideal lift-off point.
These are probably all hand-stitched though, with photo manipulation tools, and not programmatically generated, since it seems to be a one-off project inspired by the surrounding environment and terrain. It's likely possible to program a utility that creates these, as a follow-up phase to the general concept.
> The transition from ground view to aerial view gives no hint of a full-sized aircraft's ideal lift-off point.
Especially the one of the Blue Mosque (not in the original article, but from a collection it links to.) That's what convinced me it had to be a drone, at least for the oblique shots.
Yes, and you can even get similar results (of room-scale objects) by mis-using the panorama photo feature in the iphone camera app. Presumably if you strapped it to a drone with a rotatable arm, you could get something like this.
I had some tears ago the idea of using this proyection for navigation maps in cars. This was when inception came out of course.
The idea was adding a new kind of display for your TomTom in your car windshield. Obviously you need a very powerful hud-augmented reality generator to display a map like this in your windshield and don't have the image block your actual road view. Way too out of current technology and my capabilities.
We'll have to wait some more years for something like this.
Because of our brains' instinctive perception of the trap, maybe.
We automaticly find ourselves wandering around the surroundings pictured in these works, but we can't really wander, because it's like a hamster wheel (or let it be a hamster cylinder) - there is no horizon, so it restricts our ability to dream about & expand the fantasy world. No freedom for our imaginations there.
Thus, it's actually a good thing to feel uncomfortable. It tells about our eagerness for both physical & imaginative freedom, IMHO.
Could creating automated software for this be a potential computer science master thesis? I know this would be alright for a bachelor and probably not substantial enough for a PhD thesis. But I'm wondering if it is academic enough for a master thesis.
I might consider to switch, since what I'm doing now is not a project I believe in and I'm only doing it to get the paper.
As a student, I would find that an interesting thesis. As your advisor I guess.
I'm working on a generalization of this concept to rendering, but I'm not sure I would publish the results or just write informally about it. The idea is to generalize light rays to depend on the geometry of the scene, allowing them to bend to reach more interesting parts (behind objects or the horizon) as a global function of the scene configuration.
Sounds really interesting - post it on HN whenever you publish?
I like the idea of VR/AR user interfaces where light rays bend interactively according to controller/gaze position. So for example by looking at the edge of an object the light rays bend so you see behind it. Or by looking off into the horizon the entire landscape folds over like in these photos. Possibly vomit-inducing and useless, but I still want to try it.
I've always thought racing games should take this perspective, which would fix the problem that (in first person view) your most important information (how does the road ahead curve?) is concentrated in a small part of your screen and hard to see. This would combine the utility of a minimap of the road ahead and the nice overview you get going downhill in a single view.
I've already made 2 prototypes of it (a race-track on the inside of a sphere), but as always, didn't get very far. There's probably a more optimal projection to be found than a sphere, but it's an easy start.
Clever idea, would probably work well for an arcade style game if you're set on a first person view. For a "driving simulator" type (Forza, for example) game, first person is used for a more authentic experience, in which case, you may as well turn the minimap off anyway and learn the course the way you would in real life.
28 comments
[ 2.8 ms ] story [ 83.6 ms ] threadThese are probably all hand-stitched though, with photo manipulation tools, and not programmatically generated, since it seems to be a one-off project inspired by the surrounding environment and terrain. It's likely possible to program a utility that creates these, as a follow-up phase to the general concept.
Especially the one of the Blue Mosque (not in the original article, but from a collection it links to.) That's what convinced me it had to be a drone, at least for the oblique shots.
https://youtu.be/dG22TcpjRnY
https://youtu.be/z5cV-N8xDgI
https://en.wikipedia.org/wiki/Stanford_torus
http://i.imgur.com/YbPQn56.jpg
We automaticly find ourselves wandering around the surroundings pictured in these works, but we can't really wander, because it's like a hamster wheel (or let it be a hamster cylinder) - there is no horizon, so it restricts our ability to dream about & expand the fantasy world. No freedom for our imaginations there.
Thus, it's actually a good thing to feel uncomfortable. It tells about our eagerness for both physical & imaginative freedom, IMHO.
I might consider to switch, since what I'm doing now is not a project I believe in and I'm only doing it to get the paper.
I'm working on a generalization of this concept to rendering, but I'm not sure I would publish the results or just write informally about it. The idea is to generalize light rays to depend on the geometry of the scene, allowing them to bend to reach more interesting parts (behind objects or the horizon) as a global function of the scene configuration.
I like the idea of VR/AR user interfaces where light rays bend interactively according to controller/gaze position. So for example by looking at the edge of an object the light rays bend so you see behind it. Or by looking off into the horizon the entire landscape folds over like in these photos. Possibly vomit-inducing and useless, but I still want to try it.
I've already made 2 prototypes of it (a race-track on the inside of a sphere), but as always, didn't get very far. There's probably a more optimal projection to be found than a sphere, but it's an easy start.