Skip to content

iPhone RTI

14 May, 2012

I have been spending quite a lot of time lately working with Reflectance Transformation Imaging (RTI). It is a computational photography technique developed by Tom Malzbender at HP labs. It is based upon photographing an object from a fixed camera position, in each image the lightsource is moved. These images are then compiled into an interactive image within which the light can be virtually moved. As well as light movement the RTI allows the user to control a range other other variables and to render the surface using a range of algorithms which alter the surface appearance of the objects in the scene. In archaeology this technique has great potential as a means of dissemination (this beats a normal photo) but also as a means of interpreting things which are not normally visible to the naked eye. It is this technology which first inspired our (Nicole and I) project, Re-Reading the British Memorial. Much more can be found on the website of Mark Mudge and Carla Schroer’s organisation Cultural Heritage Imaging. They have popularised this technique for cultural heritage in the USA and have created an abundance of materials and advice all available at no cost. Amazing.

One of the things which makes RTI capture so fantastic is that you need very little equipment to get recording. As if to prove this point, Graeme noticed this morning that the IOS 5 version of the interface for the iPhone has a focus and exposure lock. This means that the iPhone camera can capture RTIs! Here is how we did it, with preliminary results.

So the first challenge was to attach the phone to a tripod, this was easily solved with a few elastic bands. Next and perhaps most challengingly we had to figure out how to adjust the exposure control so that the vast majority of the light in our images came from our mobile light source instead of the ambient light in the room. There are no explicit settings for this on the iPhone so you need to trick it. We put the light source right by the object we were capturing to slightly over-expose the image. We waited for the autofocus to adjust to the high light levels, then I turned the autofocus/exposure lock on by holding my finger on the screen for a few seconds. When you take the light source away you have a totally dark image which becomes beautifully exposed when your light is switched on.  Perfect.

Next, you capture as normal. We had no way of triggering a flash from the iPhone so we used a constant light source (a deconstructed desk lamp). We didn’t have a remote control either so we had to just very gently tap the shoot button and hope that the camera didn’t move. The highly sensitive controls on the iphone made this possible but I have noticed that a remote control is available so probably is worth the investment. In this case we were lucky and the camera didn’t move.

The result was extremely impressive, the iPhone camera is  only three megapixels so this is not the highest resolution option but it certainly has the power to reveal things which are invisible to the naked eye. More importantly it places the technology needed to make RTIs into the hands of anybody with an iPhone a desk lamp and a reflective sphere.

I am sure this is possible with other mobile devices too. Get your phone out and have a look!

Lit from the front:

Lilt from the front

Raking Light:

Raking light

Specular Enhancement:

Specular enhancement

Diffuse Gain:

Diffuse gain

— Gareth

I have added this post to my own blog also: Here I write mostly about my own research into Roman painted statues and digital representation, so take a look there if you are interested in following this work.


RTI Example – St George’s Church, Portland

16 April, 2012

We’ve been looking into how RTIs can be best shown through the web.

In the same way that a video needs to be opened inside a video player, to view an RTI as an interactive experience, an RTI viewer is usually installed onto the computer of the person who wishes to open the file.  But there are some great web-based options being developed that will allow users to view RTI files within their internet browser.  There is one that Hembo Pagi has made that allows for viewing RTIs within a WordPress blog, and we are certain that more versions will be developed.

For now, we have recorded a video of Gareth using an RTI viewer on his own PC, so that you can see what kind of thing you can expect from an RTI.

This is a gravestone that we recorded whilst visiting St. George’s Church on Portland.

Halfway through the video (go straight to about 37 seconds for the best bit), you will see that Gareth zooms in on part of the inscription, and switches on an option for ‘specular enhancement’, and this is where the real magic happens.  All of the text inscribed on the stone suddenly becomes visible, and easily readable. This is exactly the sort of result that we think will contribute to interpretation of problematic gravestones.

CAA2012 Conference Presentation

16 April, 2012

On the 27th March 2012, we presented the Redicovering Our Churches project at the Computer Applications in Archaeology 2012 conference at the University of Southampton.  We’ve uploaded a copy of the powerpoint presentation into Slideshare.  It doesn’t have a copy of the script, just the slides, but we thought it would be useful to share the content with you all.

St George’s Church, Portland

2 April, 2012
St. George's Church, Portland

St. George's Church, Portland

We recently visited St. Georges Church on Portland in order to demonstrate RTI techniques to a team from Wessex Archaeology who were working to document the church and its associated cemetery, you can read more about  work that they are carrying out with the Churches Conservation Trust on our Other Projects page. The project held particular personal interest to us as the church had previously been the parish church for several generations of Nicole’s family.

The church, now managed by the Churches Conservation Trust represents a unique example of 18th century church architecture situated within the quarrying landscape of Western Portland. The cemetery represents a unique documentary record of the families and communities that lived in the area during this period with many names present in the cemetery still known on the island today.

On the first day we had brilliant sunshine, which from a photographic perspective was not ideal. I feel sure that our joy at the onset of heavily overcast and rather chilly weather was not fully shared by the teams of archaeologists and conservators who were also on site at the same time. We managed to gather some spectacular results, many from grave stones which were now illegible to the naked eye, even when exposed to conventional raking light.

We hope very much to be able to revisit the church in the near future and to conduct a more comprehensive campaign of documentation and to ensure that those stones which are no longer legible might be recorded and interpreted adding to our understanding of the history of Portland and its people.

~ Gareth Beale

Branscombe Visit

21 March, 2012

The Branscombe Project

We travelled across to Devon in the beginning of March to meet with a great community-based team who have been working to collect and then share histories from Branscombe. They have a fantastic website, which we highly recommend if you’re interested in setting up your own local history website and want to know what kind of content you could offer tovisitors. Their website is full of fantastic oral histories and well-researched information about a wide variety of topics from a postcard archive, to the wreck of the MCS Napoli, to a parish registers database.

Meeting the group

The group heard about us through a mutual friend, and we arranged to drive down to talk to them about the Community RTI project that we are starting up. We couldn’t have wished for a nicer first audience for our RTI introduction! There were quite a few volunteers waiting patiently for us in the church car park when we arrived *a little* late after taking a *slight* unintentional detour on the way to the village.

St. Winifred’s Church

Setting up

Hembo speaking to the group about RTI.

St. Winifred’s Church is beautiful. With a gorgeous Norman tower, and lovely neat interior. We all sat inside and Gareth, Hembo and I started up the laptop to show the group an example of RTI, and to explain what we were hoping to achieve with the Community RTI project. The group showed us the work that they have done so far on the graveyard, which takes the form of a detailed and extensive folder of notes and photographs, drawings and plans, of the older side of the graveyard. They are planning to do the same for the earlier portion of the site, and we are going to return in the Spring to carry out a survey with them.

The First RTI

Everyone was keen to get started, and as the weather was perfect (cloudy, with no bright sunlight, but mild), we all went out into the graveyard and the group chose between them a gravestone that had details that were difficult to make out with the naked eye. As we all munched on a Hobnob (other biscuit brands are available) kindly provided by Rose, Hembo talked us all through the set up to RTI a gravestone.

Discussing options

Gareth, Hembo and I talking to the group about the project.

The Stringman

John bravely volunteered to be Hembo’s string man. It’s an important job as the ‘stringman’ makes sure that the portable flash is always an equal distance from a central point on the gravestone, no matter what angle the flash is being taken from. This is done by tying one end of a piece of string around the flash, and the other end around the stringman’s (or woman’s) index finger!

Taking images

Hembo and John doing the first RTI.

If you’d like to know more about the process of RTI-ing gravestones, including what the stringman(woman) does, read our ‘What is RTI’ page for a brief introduction to the whole process.

The Results

We reckon thinking back that it took Hembo and John about 20 minutes to record the gravestone. Then we all went back into the church to watch the ‘RTIBuilder’ software put all the photos together into one interactive file. The results were great! The RTI was clear and crisp, and the words all clearly visible and readable. The group were happy with the results, and so to celebrate, we had some more biscuits (this time provided by Gareth and I as we had spied them in the supermarket that morning and they looked too delicious not to take along with us).

Seeing the results

The group studies the results from the first RTI.

What Happens Next

We are going to be collecting some more RTIs from St. Winifred’s, with the Branscombe Project team, when we visit again in the Spring to do the graveyard survey. We’ll keep you posted here on this website as plans develop. In the meantime, I cannot recommend more a visit to the Branscombe Project website, it really is an absorbing read, but I warn you, once there, you’ll find yourself reading all the stories and will lose an hour or so easily, so prepare yourself for the experience with a cup of tea, and… you guessed it… a biscuit!

With Thanks To…

With thanks to the Branscombe Project, and also to Rose Ferraby and Hembo Pagi for helping with this demonstration, and for letting us use their excellent photographs in this article.

St. Winifred’s Church, Branscombe

8 February, 2012

We’re very excited because we are going to Devon!

In March we will be visiting sunny Branscombe to help with a project to record the standing archaeology of St. Winifred’s Church.

The Branscombe Project’s extensive website has loads of fascinating information about the village, and surrounding landscape, of Branscombe:

Our contribution to the project will be to record the gravestones of part of the graveyard to add to the excellent work already done by the team.

We’ll be using RTI over a few days to record the gravestones at the church.