Augmented Reality at Cultways

Yesterday Arc-Team attended the workshop "Cultural tourism and mobile technologies" organized by Trentino Sviluppo in the city of Rovereto. The meeting was related with the European project Cultways and, although we had not participated in the work (already concluded), we were invited to show some related research we did for other institutions (especially the Soprintendenza per i Beni Librari, Archivistici e Archeologici di Trento).

Here below is the poster we presented, done (like always) with Inkscape and GIMP:

The poster for the workshop

In addition to the poster, we prepared some Augmented Reality applications, to show the potentialities of this techniques in Cultural Heritage.
The first installation we did is a prototype for the upcoming exhibition that should take place in Padua in November 2014, as a collaboration between the Antrhropological Museum of the city, Antrocom Onlus, and Arc-Team. This event, called "Facce. I volti della storia umana", is the natural evolution of the Taung Project, and is ideally connected with the exhibition "Faces da Evolução" which took place in Curitiba (BZ). Both of the exhibitions are intended to be "open source", as the data, the software and the know-how has been (and will be) shared through the net.
Here is a short clip of the application, which is based on the joint used of Augmented Reality and 3D printed objects with RepRap (with the help of our friends of Kenstrapper):

The second installation we prepared is a prototype we developed for the Museum of Torre di Pordenone, in order to allow tourist fruition of part of the roman villa buried under the garden.

The third application regards a pilot project we are working on for the Archaeological Office of the Provincia Autonoma di Bolzano Alto Adige/ Autonome Provinz Bozen Suedtirol. The research is connected with a survey to map and document the WW1 evidences in the territory. As an experimental stage of the same project we are testing some Augmented Reality tools to develop cultural sightseeing of the landscape, looking to the commemoration of this event which will be in 2014/2015. Currently we are considering the possibility to print interactive maps like the prototype below:

Just an example of interactive map

Please notice that the informations displayed in the map are not geographically correct: they are just an example to show all the possible kind of data which can be loaded (images, movies, 3D models).

Another possibility for tourist fruition are interactive panels, like the one below...

A test for a panel

... in this case we just added a simple image, but, like the map before, we could load movies or 3D objects.

All the Augmented reality applications were done with Open Source and Free Software. I will describe these tools in the next post, but you can already find some information here in ATOR (just search for this topic).

III Meeting of Forensic Dentistry and Forensic Anthropology at University of São Paulo: The triumph of Free Software

Upper part of the panel about photo scanning with free software, awarded first place among all submitted to the Journey .

Few people have closely followed my studies in forensic facial reconstruction (FFR). These individuals not only witnessed the positive occasions of institutional acceptance, published in this website and social media, but also saw the criticism and even some attempts at discouraging my quest to develop a new methodology in this area.

From the beginning I counted on the support of archaeological research group Arc-Team and archaeologist specialized in Egyptology, Dr. Moacir Elias Santos. Although, I lacked a partnership in the field of Forensic Dentistry, something crucial for this type of research.

Panels made ​​with Inkscape and exposed at the III Meeting of Forensic Dentistry and Forensic Anthropology at Faculty of Dentistry of University of São Paulo (FO-USP) and the award for 1st place.

By teaming up with Dr. Paulo Miamoto, PhD student in Forensic Dentistry at the FO-USP , I found the strategic and prolific partnership I was seeking.

A few months ago, I received a call from Dr. Rodolfo Melani, head of the Laboratory of Forensic Anthropology and Odontology (OFLAB - USP) inviting me to attend the III Meeting of Forensic Dentistry and Forensic Anthropology at FO-USP.

As I wrote in my weekly newspaper column, receiving an invitation to lecture at USP had the same emotional effect as an Oscar nomination for an actor. So I readily accepted the invitation.

In addition to lecturing at the event, I also would teach a pre-meeting course in Principles of Forensic Facial Reconstruction with Free Software, with the aim of presenting open software to this market niche, consisting of dentists, forensic experts and other professionals.

The class had professionals from various Brazilian states: Sao Paulo, Rio Grande do Sul, Rio de Janeiro, Paraíba and Rio Grande do Norte. Among these professionals, there were professors, forensic experts from state police departments, federal police department, undergraduate and graduate students.

Course Principles of Forensic Facial Reconstruction with Free Software - School of Dentistry, USP (FO-USP)

Dr. Miamoto and me taught a syllabus featuring open software like Blender, where we approached basic 3D modeling; InVesalius, in which we explained the basics of visualization and reconstruction of CT scans. We also presented 3D file editor MeshLab and 3D model maker from photos, the nice and new PPT-GUI.

Lecture on Digital Facial Reconstruction featuring the exposition Faces of Evolution.

The Meeting begun the next day, September 26th. The lectures were fantastic, many of them presenting very nice technologies for investigations. I could watch some and was happy to see that Gimp and other open software were cited in the speech Dr . Luiz Eugenio Nigro Mazzilli, a great statistician, professor FO-USP .

Dr. Miamoto talked about FFR with Dr. Silvia Virgínia Tedeschi-Oliveira, the developer of a nasal projection estimation method in Brazil, from data taken solely of the skull.

In my lecture about Digital Facial Reconstruction I could tell some of my experiences with this science, explain about free software and tell some interesting facts that occurred during these studies.

Dr. Paulo Miamoto presenting a panel on 3D scanning from photos with free software

In the event there was also a presentation of panels, in which I took part by making several of them using Gimp. However, I was quite pleased to see that there was another panel describing a technique of Dentistry, with the use of free software .

Panel presented by Dr. Thiago Beaini (not Arc-Team), where we see the use of InVesalius and MeshLab .

The panel in question was presented by Dr. Thiago Beaini . We can see that he used InVesalius and MeshLab to accomplish the research .

Dimensional deviation graphic generated in software CloudCompare

In addition to the studies presented in the panels, I noticed in a post on the Facebook webpage of the event, that Dr. Thiago Beaini was also using CloudCompare, another open software widely used for the generation of 3D deviation graphics.

Panels composed by me and my research partners were all made with Inkscape and Gimp .

Above we have a panel on a scanning photographs, which was awarded first place.

Panel on a blind test of a forensic facial reconstruction.

Panel on forensic facial reconstruction of Australopitechus, made in partnership with archaeologist Dr. Moacir Elias Santos .

Panel on comparison of manual technique modeled by Dr. Miamoto, versus digital made ​​by me.

Final words

Personally, it was a great honor to participate in this event, marked by excellent lectures, a powerful networking and good memories that gathered the best of scientific knowledge with the finest of interpersonal camaraderie.

I am happy to have contributed with my knowledge of 3D computer graphics linked to free software. I imagine the freedom seed was planted in the minds of students and professors of one of the largest and best universities in Brazil.

I hereby acknowledge Dr. Paulo Miamoto for translating the original post, in Portuguese.

A big hug to everyone!

OpenJUMP: adding a vector layer with a customized db

Hi all,

today I recorder a fast videotutorial regarding the GIS OpenJUMP (which is one of the Geographic software integrated in ArcheOS).

Normally, when you add a new vector layer in your project, you get a new empty level, without any database schema, so that you can start to draw your feature, but, if you want to add some info, you have to manually describe your db schema, like in the video below:

Of course, if you have to draw many different layers with a common database schema (like always happen in a normal archelogical GIS), this operation can be time-consuming (and boring). For this reason in the videotutorial below I try to show how to write a short script which automatically add in OpenJUMP a new vector layer with a customized database schema:

Since I fear the quality of the video is too poor in Youtube, I prepared an image in which you can see better the source code of the script:

The code of the script

To work correctly, the script has to be placed in your OpenJUMP folder, in /lib/ext/BeanTools/ and, as you see in the video, you have to refresh the menu in OpenJUMP (Customize --> BeanTools --> RefreshScriptMenu) to find it (in ArccheOS Caesar you will find the script already in the menu. Just modify the code according to your needs).

Like always I added the tutorial in our ArcheOS wiki (DADP project), in order to go on in composing a free documentation system for Digital Archaeology: 

• Tutorial 1
• Tutorial 2

I also uploaded the code of the script into a specific github repository, so that, if you want, you can contribute in its development. We can use the comment space of this post for the discussion about the schema and about its possible modifications (or you can simply download the script and modify it in order to fulfill your specific needs).

I hope it was useful, have a nice day!

File conversion with TOPS (Total Open Station)

HI all,

today I had to convert some raw total station data from Trimble file format (.are) into something simpler and more readable, like a csv file (Comma Separated Value), in order to further process my points and load them inside a GIS. To perform this operation I used a specific software which you can find in ArcheOS: Total Open Station. 

This tool is developed by +Stefano Costa and Luca Bianconi. Here you can find the official website.
I record a short videotutorial about the file conversion. You can see it in the video below.

As usual, I uploaded it also in the Digital Archaeological Documentation Project of Innsbruck University. Here is the direct link.
I hope it will be useful.
Have a nice day!

Scanning a face in 3D with photos

Scanning a face usually is a problem, when you use SfM (PPT-GUI) to make this. Maybe because of the color be almost the same around the surface of skin, the algorithm doesn't get to reconstruct the face entirely.

Some days ago, a friend of mine sent me a sequence of photos of his face. I knew that it did not works, but I tried even this way. Like I knew, the face doesn't reconstructed entirely, but the T-shirt of my buddy yes. Even the fabric having the same color in almost over the surface.

I thought that, maybe, if I colored the face before scanning it, the surface would be recognized by algorithm. So, I took to seqeunces of photos. One without traces over the skin, and other with traces over the skin.

You can see in the scanning above, some areas empty.

Tutorial: Howto make 3d scanning with picures: http://arc-team-open-research.blogspot.com.br/2012/12/how-to-make-3d-scan-with-pictures-and.html

 

But when the traces was done, the scanning worked a lot better!

Well, the problem of the surface was solved, but, and the texture?

Well, I could use two photos, taken in front and side, but the face would be deformed by the perspective... I don't have a good camera (with controls) and this is expansive here in Brazil. I had to solve this problem using my simple smartphone with a 3.1 megapixels camera.

Fortunately I could use the data of the first scanning made without trace the face.

Addon: Bundler file importer (.out): https://projects.blender.org/tracker/index.php?func=detail&aid=30893&group_id=153&atid=467
 

Initially I imported the bundler file (.out) inside  Blender (thanks to rgaidao's tip). The bundler file have the cameras with photos and a sparse points cloud.

So I imported the reconstructed face and I aligned it with the sparse points cloud.

 

Using four photos I projected the texture over the surface.

After that, I created new objects to complete a scene and render the final image.

gdal_edit.py: if you can think it, you can do it

Hi all,

today I'll write a post about a python script I used to edit the projection of an old geotiff, to import it in the GIS I am managing for Aramus archaeological mission 2013. To understand why I needed to use such a script, I have to explain briefly the problem occurred this year.

Since eight years (2006), Arc-Team is supporting the excavation in Aramus (AR), currently leaded by Dott. Walter Kuntner of Insbruck University (Institut für Alte Geschichte und Altorientalistik), in colaboration with Prof. Hayk Avetissian (Yerevan State University). In 2006 Ing. Klaus Kerkow and Christine Hanisch produced an high resolution aerial map of Aramus hill, using S-42 geodetic datum (with the Krassowsky ellipsoid). Unfortunatley, since 2004, Aramus project has been set up on a modified version of S-42, to avoid bureaucratic problems. This is the reason why I had to convert the aerial map geotif from the Krassowsky-Pulkovo to the local Aramus system. If I could work with a georeferenced raster and a plain world file, it would have been simple to solve the problem, just modifying the coordinates directly in the code... but working with a geotiff, things became more complicated. I could georeference again the aerial map, but I wanted to spare time and keep the same quality of the original raster. Luckily Walter Kuntner found the right solution in internet: gdal_edit.py, a simple scrit able to "edit in place various information of an existing GDAL dataset (projection, geotransform, nodata, metadata)". I checked quickly If I had this script in my GDAL files, but I did not find it, so I simply copied the source code from the OSGeo SVN repository, at (this link) and I saved it in a .py file. Than I get the coordinates of the upper left and lower right corner of my geotiff in OpenJUMP ...

Copying the upper left coordinates of the geotif

... and finally I modified the GDAL dataset with this command:

python gdal_edit -a_ullr ulx uly lrx lry

where ulx and uly stand for the new x and y coordinates of the upper left corner of the geotif, while lrx and lry are the new x and y coordinates for the lower right corner.
I know that this problem is not very common, but I hope that this post will be anyway somehow useful to ATOR readers.
Have a nice day!

OpenJUMP: query to extract single points from a general vector layer

Hi all,

today I am working on data elaboration of an archaeological excavation. I decided to record a short videotutorial to show how to perform a very simple query in OpenJUMP (which is a topic I am often asked to explain during lessons about archeology and open source). Before to start, I'll write a short introduction regarding the processing of this data, just to understand why I need to perform this kind of query. 

I worked on this excavations collecting all the data in a local system (with simple 3D Cartesian coordinates), because the job was an emergency archeology project and I did not have the time to set up a geographic coordinate system on the field, before the construction site began. Just at the and of the excavation, I could come with a RTK GPS, to collect some Ground Control Points (GCP) in UTM WGS84. This is the reason why in this time, when I am processing the data, I had to put together all the daily total station downloads in a single cvs file, which I georeferenced in OpenJUMP, using the GCP I collected with the GPS. The problem is that now I will have to separate again the single points, grouping them according to their function (e.g. points for photomapping of area 1, height points of the same area, points for SfM georeferencing, find-points and so on...). To do this operation, I will simply use OpenJUMp (where I loaded and georeferenced the csv file), performing a query on the attribute (the name) of cumulative vector layer , which will tell me where are the the points I need. In this way I can select in the GIS the features I am looking for and put them in a new vector level, copying the db schema of the cumulative layer (name, y, x, z, code). At the end of the process, I will have single separated vector files which I will use in the next steps of data processing (photomapping, 3D, ecc...).

Here is the videotutorial I uploaded in the ArcheOS tutorila wiki (rom Insbruck University):

I hope it was useful, have a nice day!