Paul Coulton's Forum Nokia Blog

How best to present 3D navigational information to pedestrians?

2008-11-28T11:41:25+02:00

A few blogs ago I said we had created a model of the Lancaster University campus which we (myself and Carlos) would transfer to the mobile phone to test some 3D navigation ideas for pedestrians. We have managed to get the model into M3G (and working reasonably well despite not having yet optimised the code) and the next step is to build in GPS (and eventually Compass) data which will control a you-are-here globe which will move in real-time.

This extra data along with the colour selection of the buildings already seen in the video will allow us to explore how best to present 3D tour navigation to the user (for example first person, third personand, fly through etc) what type of control of the view and functionality is optimal for the user for a particular requirement i.e route navigation, POI information, etc . I’m hoping to build on my favourite paper ‘Exploring the design space of Smart Horizons’ form this year’s mobile HCI which looked at user viewing preferences of 3D spatial data and POI. I will keep a record of this project progression going on the blog so perhaps the community can pitch in with any thoughts or ideas which will help us evolve the project.

A Zero Button Game

2007-09-07T15:48:19+03:00

I have been meaning to write this blog since the Champion day in Singapore (yes I know I am slack) as it sprang from a conversation I was having with Kevin Sharp over dinner.

Basically, we were talking about mobile interfaces and in particular the desirability of the one button game but when I was showing him the game my student Paul Gilbertson had just completed we realised it was in fact a no button game.

The game itself is called Tunnel Run and is designed to run on one of my favourite development phones the Nokia 5500. Those of you in the know will realise that this is the device with the in-built 3D accelerometers that I am so fond of. It builds on the Poppet framework we created for Tilt Racer and Paul has used this to create a Novel 3D driving game. The idea was to test the validity of ‘tilt’ as a valid input for mobile gaming we would not create a game that traditionally uses tilt as an input sucj as a marble game.

During the trials, we recorded the users’ comments and then got them to asses the merits of the tilt interface compared to the traditional keypad interface for game playing. One of the most usual comments when the users first started the tilt game was “that’s so cool” and “this is a lot more fun than the other one”.

The unanimous opinion was that the tilt interface was the most fun and they all preferred it to the traditional joypad and a number of the users said it was “easier if you have big fingers”. Further, all of the users identified the problem with the tilting the game forward as a game design issue rather than an interface issue and were very impressed by the granularity of tilt control which appears to have been a significant problem when using the camera.

Interestingly a number of users asked whether “this was a commercial phone” and “whether they will put the sensors in all new phones” which, along with the comment “this is great I don’t usuall play these types of game” should be encouragement to manufacturers to consider the incorporation of the technology to a wide range of devices.

You can see the game in action here and if more phones appear with 3D accelerometers maybe we will publish it one of these days.

Create your own 3D photos using your mobile phone

2007-05-25T18:15:26+03:00

This is definitely one of those ideas that I wont even pretend that we had some serious academic motive behind its development, and it definitely comes into the category of doing something ‘just because you can’ (but not necessarily should). It started as conversation between Will and I had about doing something fun for an undergraduate project but in the end we just couldn’t wait. Essentially it’s a J2ME application that allows you to create your own mobile 3D anaglyph photos using the mobile phone camera by taking two slightly offset pictures with the result separated into red and cyan colour planes as shown and in a video HERE.

The software allows you to then adjust the relationship between the red and cyan planes so you can get the best image to either view on you phone or up-loaded to the newly created site.

Of course you need to have a pair of 3D specs to view it properly but you may have some from either a trip to 3D movies or you can pick some up cheap online. We managed to get some rather natty Shark Boy and Lava Girl ones as shown. Perhaps I could persuade Anina to make them this years fashion must have J

The application is really quite fun and addictive as it takes practice to get really good.

Will has made the application available on the site in a number of different versions so download and enjoy J

Mobile Location Mashups

2007-05-23T15:32:20+03:00

Someone told me the other day that I had been suspiciously quiet of late so this post may well explain why.

We (me an Will Bamford) have been trialling a new research project called LocoMash which allows communal photo mash-ups of events to be recorded using mobile phones with either external or internal GPS (Nokia were very kind and lent me an N95 J) and then displayed on the web on Google Maps.

I video of the application in action is available on you tune HERE.

The idea sprang from Mass-Observation which was/is a social research organisation in the 1950's whose remit was to record everyday life in the UK principally through direct observation. In this project we aim to create novel tools and systems based around mobile phones to enable everday life and events to be recorded in a richer way than ever before directly by the people involved.

We used the location API to create zones which are downloaded at the start of the mash-up and displayed as part of the user information. At the website these zones can be used to filter the images spatially. If you follow the link to the site above and select a particular event then select all users and all zones, before you hit GO ,you can see the photos from the whole event.

We have also allowed the users to define their own contextual tags at the time of blogging to allow more contextual based searching.

The time space blogs can stored on the phone until the user decides to upload them via GPRS or WiFi which will hopefully make it versatile.

If anyone is interested in using the application and site to record an event, and don’t mind us using the data for research papers, I will happily discuss this especially if its in sunny climes and you want us to attend ;).

Mobile phones as data loggers and sensors

2007-05-10T13:02:01+03:00

I previously posted an outline of a project I am working on with Will and colleagues from our Geography Department (Duncan Wyatt, Colin Pooley, Marion Moser, and Gemma Davies) which uses mobile and GPS units to both track (recording at 1 second intervals) the journey to school of 30 local children. The Geoblog app alsoallow the pupils to blog their experience along their route to create a richer picture than the route alone. A couple of people have asked me have we got any results so here is a quick taster of what we are producing and likely I will do more at a later date. A mixed methodology is used which links the data from the phones with a modelled pollution surfaces using a GIS system along personal and interviews with the pupils.

The figure below produced on a GIS system shows the journey of 3 pupils (names being pseudonyms chosen by themselves) on their journey to school. Scott cycled while Ella and Louise walked. From this we can ascertain the relative exposure levels to the pupils along the journey. An interesting point to note is the journey of Louise who cuts into the park parallel to the road and when asked in the interview about this she said it was because she didn’t like the traffic noise but was only allowed to do this at certain times in the year for personal safety. This highlights the need for a mixed methodology approach to these types of study and the power of inter-disciplinary research.

Based upon 2004, OS Landline. Map produced on behalf of The Controller of Her Majesty’s Stationary Office © Crown Copyright. Lancaster University, Bailrigg, Lancaster, OS Licence No. 100028811.

The project is producing some very interesting results and shows what interesting things we can do with our increasingly powerful phones and their potential as a very effective sensor and data gathering tool.

could you be an urban outlaw?

2007-03-30T14:40:18+03:00

We have finally got round to arranging large scale experimental evaluation of our research project the serendipitous proximity shoot-em-up game Mobslinger.

We have therefore just put a beta version of the game on our site for testing by any one with a Bluetooth enabled phone who has a bit of time to waste with a few friends and fancy them selves as an urban outlaw. Currently it’s a J2ME version with only Bloodbath mode implemented but the S60 version will follow soon. We are going to put a league table (to enter your kill rate) on the site in the next few days together with the all important feedback form .

Although the first beta has limited game depth we would still welcome any comments from would be outlaws particularly on the following research questions:

whether its fun,

the social/work scenarios in which the played the game,

whether they played with friends or encountered strangers,

how it could be improved,

Other comments relating to probable porting issues; bugs, etc are also useful.

And to those of you who do download the game DRAW!

Using gesture as an interface for mobile applications

2007-03-12T15:13:36+02:00

Last week I was at the Game Developers Conference and heard many complaints about lack of innovation. With this in mind I gave a lecture on “Using Touch, Sight, and Gesture in Mobile Games”. Whilst I covered using RFID/NFC and the Camera on the mobile phone as a novel interface, the aspect that seems that drew most interest from the audience, and later on the web, were some demonstrations (available on Youtube: here and here) for games we (myself, Tamas Vajk and Will Bamford) created using the Nokia 5500 in-built accelerometers to control games running on a PC via Bluetooth . The original game was based on the old tron classice light riders is shown below but the videos show the newer Tilt Racer.

My proposition in the whole talk was that using Nintendo and the DS and Wii as an inspiration for attracting a non-traditional gamer audience (although some comments on web relating reports of my talk wrongly assumed I was suggesting phones as a replacement). I believe that accelerometers have an enormous potential for attracting non-gamers as Andreas Jakl already indicated and indeed atracting this audience was highlighted as the inspiration for the DS and Wii by shigeru Miyamoto in his keynote.To highlight this in a none-game setting I thought I would also add some plots of the sensor outputs to illustrate my point.

In the following we see the three accelerometer outputs when the phone was simply placed upon the desk. It can be seen that outputs x and y are approximately zero although some sensor noise is and output z is showing a positive 1 g force which is the effect of gravity on the device. Note: that this noise could be reduced by the application of a digital filer on the output values but in this case we are showing the raw data. Overall this Figure illustrates that the effects gravity provide a means of deducing the orientation of the phone which can then be utilized to provide a ‘tilt’ based controller.

In following we see the accelerometer outputs resulting from a rapid sideways movement of the phone across a desk. At the start we see outputs in the same state as in first Figure and then very large acceleration forces on y produced by the rapid sideways movement. There are some smaller residual artefacts from this movement seen in both x and z but this because the test was performed by hand rather than using a fixed jig and its difficult to isolate a movement completely under such conditions. We clearly show the potential for utilizing ‘hand gestures’ for the control of events within. However, gesture recognition requires careful study as the variation in how the user holds the phone could produce anomalous outputs and there is no method of obtaining the phones physical position within actual space.

The final figure shows the output after the phone has been dropped. The trace shows that initially the phone is held upright with the screen held at the top and gravity is acting on x. When the phone is dropped the effect of gravity is overcome and all three accelerometers are zero before the phone hits the floor with a jolt. The aim of the test was to show that the phone could be used to measure activity of a player as per the snowboarder application developed by Andreas and his students.

Hopefully the games and applications already developed highlight the potential for developers but I would also urge Nokia to add these sensors to as many of there devices as possible as having a critical mass of devices in the market will be essential for getting operators to take this seriously.

Mobile health blogging on the journey to school

2007-02-20T11:16:12+02:00

I have been a little quiet on the blogging front lately as Will Bamford and I have been working hard to complete the first stage of project we are working on with members of our Universities Geography department.

We have created a version of Locoblog, which we have christened geoblogger, that will allow us to record the journeys to school of 30 children during the course of a week of their journeys to and from school.

No only do we record the route but we have given the children the ability to blog their health status along the journeys together with an concerns about the environment.

The application also provides a twice daily health survey which is sent back to us along with the distance travelled.

We are going correlate their children’s school journeys with local traffic pollution data in the hope that we can make recommendations on the best routes to take and whether the local council needs to consider creating green corridors for such journeys. Here is the kit we have a amalgamated for the project which looks a bit scary all in one place .

The experiment will take place in 1 week periods in each season over the next year with the first one starting next week. I will blog more details and perhaps some preliminary results during the course of the trial as it will hopefully provide some interesting insights in using mobile phones for fieldwork recording and experiences of using this technology with children

Nokia integrated NFC phone

2007-01-12T11:27:25+02:00

While the main press furore at CES may have been around the iphone I was much more interested to see Nokia announce its first fully integrated NFC phone based on the Nokia 6131. The specification says that it supports JSR 257 but doesn’t list this of those preinstalled which may mean you need to get access through Nokia as for the previous NFC phones

The phone will allow users to have their credit card, travel card and loyalty card in on a phone plus many other activities. Much is talked about disruptive technologies, i.e technologies than produce enormous changes in society, and I believe NFC could well be such a technology particularly when coupled with a phone. As and example I blogged a while ago about the NFC poster specification being finalised and coupled with this type of device we could well be at the dawn of a whole new advertising and distribution were a poster for a band may allow you to download a ringtone or a poster for a movie may allow you to download the trailer.

So I say forget the iphone get yourself a Nokia 6131 NFC!

When hand waving becomes a dangerous sport!

2006-12-18T15:57:19+02:00

I have been following the growing number of wii related accidents on the aptly named, wii have a problem, which ranges from black eyes and cut hands, from inadvertently hitting light fittings or friends, to broken monitors if the straps on the controller break.

This is interesting to me as we are just finishing a project called Poppet which allows the 5500 motion sensors to be used as a PC controller via Bluetooth. Perhaps I need to get some better accident cover before I test this!

PS: Has anyone got any details of how to get the J2ME drivers working. We have done the project for S60 and have got a J2ME version working on the emulator but cant get it to install on the phone.