Augmented Reality (AR) - Computerphile

Key Takeaways

there's a lot of sty over substance within AR or or any Innovative technology like VR as well why are we doing it is it going to benefit the user in long term um or is there another way of doing it that's that's equally easy um that they're more familiar with and they're more likely to to interact with so we've got an open kind of generic Dell computer in front of us now I have created a 3D tracking model and using augmented reality we can superimpose information in this case it's the Riser card cage that it shows um the user how to pull out you have to flick the switch at the bottom and pull that out if you need to replace any of the components within the motherboard so I've just opened up the AR app and you can see that it snaps to the different elements in the 3D object we have in front of us and you can see at the bottom it has an animation about how a user would go about pulling out the Riser card cage at the bottom and this can be a linear experience it can be an experience where the user has control of different use cases where they'd have a variety of different computers in front of them so people are already using this yes it's already being used kind of in industry and it's starting to be used within education so you've got the AR layer which essentially is initializing the The Experience bringing up the models Etc and and then you can add an HTML 5 layer on the top of it which can incorporate um more fun user fun lead functionality such as being able to use gestures and freezing the screen that sort of thing and that's using JavaScript based library on top of the AR environment so it kind of adds a bit more value to the AR experience rather than just showing something pretty it allows the user to interact with it and get what they want from it this is like using image recognition so a simple image that's been printed out and this is a small area of grenwich in London and the information originally came from quite a large data set surrounding geospatial information in the area of grenwich in London and through kind of 3D mapping we're able to interpret the land use um whether it's residential and which business areas the area of veget ation around the mapping itself we can switch to different layers um this gives the user quite um interactive potential in that they can turn the gestures on they can scale it up they can turn it around they can move it away they can freeze the tracking which I guess a lot of students and users will want to do because holding up a device isn't really great user experience yeah it must be hurting your hand holding it up there it is that surely one of the issues with doing this is holding a device in your yeah um one of the issues is you know having to hold it in front the environment or the image but you can overcome that by using functionality that you can employ to freeze the tracking and then you're able to take it away you may have a group of friends that you want to interact with an image and it's a lot easier um being around um a table looking at a static image where you're not forced to to hold the the iPad device up in front of you so being able to take it away from that initial tracking and then messing around with it and rotating it and and doing loads of things with it so we've got here a print out and you're pointing a device at it what's the difference between just not having the print out in the first place and just having the device what why would you use AR in this instance rather than just I guess it's more of a immersive experience it engages the user a lot more because they're part of that experience whereas um if you imagine a lecturer in a theater um speaking to a large class they're quite disconnected from the knowledge transfer here they're part of this learning experience they're engaging with something that they can see it's very visual they can interact with the environment in front of them more than they would if they were just taking notes or visiting a website or even um messing around with a cad model in a piece of software I've got to ask a question is it Sur this is is a gimmick isn't it um I think in a lot of cases and I think that's one of AR and any new kind of innovative technology shortfall is that the the money tends to be surrounding marketing and advertising and that's what is projected into the mainstream um media so a lot of the detractors will say yes it's great this this advertising campaign is great for the first time that someone watches it but what is the the worth past that experience users will or visitors will not really go back to it in the future so I think when I create anything to do with ar it has to be quite immersive but it also has to have quite a a long shelf life users have to return to it and gain valuable Insight which they couldn't do in a 5c advert like any Learning Resource it needs to be something that's less of a gimmick and more tangible in terms of a good use of their time because students are probably the biggest critics when it comes to any new technology they will immediately switch off if they can't see the value in it or if it's replicating something that they could already get quite easily Hollywood is depicting one version of AR we watch something like Iron Man or whatever and you're going to have you know heads up displays and all this sorts Stu going on how close is that to reality things like Microsoft's new announcement any sort of IT company are always trying to push the boundaries and it's about managing expectations I think I think if the users have an expectation that they can go into a Hol deck like star Star Trek generations and be completely immersed it's probably not realistic and one of the issues at the moment is the hardware is kind of holding back the Innovation so there's great ideas that are out there but it's just a hardware for instance 3D models anything over about 20,000 polygons in a 3D model and it will crash the app and that's detrimental to to Main adoption I think so if you had um a user experience where the user is going to become frustrated it's not going to work effectively I think we're a long way from what the hollow lens predicts will be something that we'll interact with on a daily basis you mentioned a certain amount of polygons so going back to that view you had there of Greenwich how many polygons is something like that showing is that I think there's about between 5 and 10,000 polygons but the original area that I wanted to deal with was um Millions so having quickly learned that that doesn't translate into a mobile a good learning experience for a mobile environment um you have to manage your expectation and if that means like limiting the amount of polygons maybe the 3D models won't be as photo realistic as you want them but in terms of the learning experience will be a lot more robust so you can create any color you like on your monitor by varying the different amounts of these three colors on a printer we work in a completely different way so we're not talking about light anymore we're talking about ink