Python Photogrammetry Toolbox for Windows 64bit

After a report of some problems about installing Python Photogrammetry Toolbox on a Windows 64bit machine, i tried to do some test on it. Finally i found the error inside a python script (RunCMVS.py). The error was reported to the developer of the software (Pierre Moulon) and he will change the original packake as soon as possible. In the meantime there is a small installation instruction to correct the error at this link.

Which GIS? gvSIG

To tell the truth, i don't use very often gvSIG. Anyway it is for sure one of the software which has evolved faster since its first inclusion in ArcheOS. Moreover the program is pretty similar to ESRI GIS, so it is perfect to help new users of FLOSS (people for whom a direct migration form ArcGIS to GRASS could be traumatic). An other important aspect is that gvSIG is maybe the GIS in ArcheOS which has the easiest (and most functional) tool to get direct layouts, so if you settle for basic maps (without to many complications) this is probably your software. In the image below you can see an example of layout model, done with an old version of gvSIG.

Which GIS? OpenJUMP

We are often asked why there are so many GIS inside ArcheOS (GRASS, gvSIG, OpenJUMP, Quantum GIS, SAGA GIS, Udig). Today I start answering with OpenJUMP. This is my favorite software for an excavation GIS, due to its fantastic drawing tools (that make it similar to a CAD). So, from my point of view, OpenJUMP is the best GIS to draw vector layers from georeferenced photomosaics, like in the image you see below...

... but of course there are also many other tools, which sometime are usefull to handle excavation data (below you see the "warping" utility). Moreover the database integration in the program is pretty enought for this kind of small projects.

Something stupid

Yesterday evening i was working on an illustration for an archaelological publication. The image reguards an amphora I modeled in Blender, which for many reasons I had to cut in half (and along different sections). I lost a lot of time cutting the amphora with a (very simple) boolean operation inside Blender, but the final result was not good enought (especially considering the wasting of time). Finally I had the idea to export the amphora from blender and import it in ParaView, to use the simple "Clip" tool inside the very large filter collection of this software. This simple/stupid solution saved me a lot of time, so I complete my work inside ParaView. It is not the first time I have to use both of these software for a project and, like before, I used the .ply format to exchange models from Blender to ParaView and the other way around. 
In the image below you can see a screenshot of the amphora in ParaView.

GNewgnewarchaeology

Good news from University of Ferrara! A new workshop about Free/Libre and Open Source Software in archaeology will take place on the 13th – 14th October 2011. More info on the official website. It will be a good oppotunity to meet and exchange experiences and informations. See you there!

GNU logo

CMVS/PMVS2 40% faster

Test data fir Nghia Ho implementation

We tested the implementation written by Nghia Ho and the results are amazing. Data elaboration was effectvively 40% faster. Now we are working on the new package for ArcheOS, wich will be released ASAP.
In the image you can see the data we used for the test: the old version of CMVS/PMVS needed 153 minutes, while the new implementation completed the process in just 92 minutes.

3D for archaeological finds

One of the most common question we receive about SfM and Computer Vision techniques is: "is it possible to document in 3D also small objects like archaeological finds?". The answer is yes, it is, and there are two possible ways:

1. putting the object to record on a turntable (with a black panel as background) and taking pictures from a fix point  (as I described in this post)
2. taking pictures walking around the object (the easiest way)

Untill now, in both cases, it is better to use Bundler 3 instead Bundler 4 (that seems to have some problems with symmetrical objects). Anyway for option 2 there is no need to "erase the background" so it is possible to use also the newest version of the sofware (which is the one inside Python Photogrammetry Toolbox). This is because other objects in the background can help Bundler to find the right position of the cameras for every picture. the image below is an example of this situation: I took the photos walking around the archaeological find, which was simply placed on a piece of wood (over a normal A4 paper, for metric references).  

3D documentation of arcaheological finds with Bundler and PMVS2

The archaeological find in the picture comes from the mission in Khovle Gora (Georgia) of University of Innsbruck, Department of Ancient History and Near Eastern Studies, (Walter Kuntner and Sandra Heinsch) and Javakhishvili Tbilisi State University (Prof. Vakhtang Licheli).