Yeah, the audio grafiti thing sounds cool. I'm sure it will get horribly abused within 5 minutes and require constant vigilance to prevent people from being insulted when they pass wal-mart, but when I saw the thread title my first thought was "Signs, definitely."

Have you heard of the Orcam? The idea is that it tries to recognize objects, but you have to record what the objects are. If this could be integrated with the cloud, so that it can recognize things that other people have labeled, that'd be neat. (Also if it wasn't $3k for a piece of equipment that would be much better integrated with other devices or software, such as the vOICe.) The result would be kinda like TapTapSee, but it could recognize whatever you're looking at on the fly because it uses a tiny earpiece camera.

Someone needs to toss a few million into all these disparate technologies and license them all into a single suite for existing platforms. You'd still have the option to use them individually, of course, but as it stands I think the vOICe is the only one that is trying to do multiple things at once, and they've been around since, what, 1998?
Having all of these things together would be much better than having to work out what you do and don't need and switch between them (which is often going to be a slow and clunky process). Like, if I could combine your app with whatever has surpassed Blindsquare, a cloud-based Orcam, TapTapSee, and the vOICe (or a more object-oriented color-based image-to-sound program, like Iodimony but better), that'd cover a ton of stuff right there. I mean, information overload and drowning out important external sounds would be an issue to address (if I'm crossing a street I'll probably want to mute everything, for example--I would not trust any app or gadget I am currently aware of or imagining as more than a distraction in that situation). But I think I'd use TapTapSee more if I didn't have to go through the process of switching and opening apps and positioning the camera and finding the lighting... so it's mostly good for identifying money, and in spite of all those testimonials I have not gotten it to identify many cans or bottles with more than color. I'm not spending $3k for an Orcam if I can get the vOICe to do most of the work for the price of an ordinary headcam, but if I can get the Orcam software, the vOICe, and TapTapSee to use the same camera at the same time, maybe even as a background app, that'd be worth... $30, at least .

Umm. I rambled again.

Yeah, I like your idea, and I agree with Jeff that overhanging obstacles and sign/fence posts that are at dangerously far-from-90°-angles are things that I and many others would very much like to know about before the injury.

Hi.
The thing with the orcam is that if you would give it internet access, it would lose it's state of an assistive technologie in some cuntrys, even the uS I think. Sure, if this thing would have had internet access, you could easily pick up things from the cloud like POI data, city maps and so on, but sadly, this will not work so easily.
Greetings Moritz.

Some cane-mounted overhead-detecting devices exist, but IMO they're not very popular for a reason, other than obscurity/price/etc. A properly mounted headcam with a vOICe equivalent would probably work better.
(Though I am given to wonder... if you had enough hair to put in a spiky style, so it sticks up and looks more querky than outright bizarre, would that provide any information whatsoever? I ask because that whisker experiment a while back got me thinking about the whole subtle-vibrations-and-air-disturbances-sensing thing. I mean, Iroquois warriors tried it with scar tissue, but no idea if it actually helped.)

As I understand it Peter Meijer is still working on a mobile raspi kit for the vOICe that uses a pair of camera mounted glasses. I think it may also includes other navigation features, the thread [here] may be well worth a read to check what they have planned and its current progress.

Crap I didn't even consider glasses LOL.

Hmmm a camera could work too but it seems a bit of an excessive solution if something like an infrared sensor could do. If it's able to calculate the distance you could even have the vibration mechanic vibrate more as you approach it. I don't know anything about building hardware though...

on the flip side, it's not a good idea to be walking about with a rather obviously visible camera mounted to a pair of glasses. When you accidentally walk into or passed an area with a bunch of kids, for example, it's my navigational aide isn't going to fly. One of the largest flaws in these items. Remember how pissed off people got with wearers of goodle glass? Those people could see where they were pointing their cameras. Imagine mounting them on people who aren't always aware of where they are looking.