Virtual Tours: Google Art Project

Hello everyone! It is a miracle I am here! My mom’s desktop has so many viruses it’s insane, because of that I wasn’t able to download flash and so what I will write about the Louvre will be from memory… unfortunately… I’m sorry (I tried).  Anyway, this weekend I read “Meaning in Motion “by Diane Favro and she argues that a lot of factors go into executing an interesting, accurate and engaging 3D model.  Academics can be limited by “availability of data, the technology used, time allotted for creation,” funding and more, not to mention the need an artistic eye. One of the projects they worked on, was actually a collaboration piece with a production company and that really helps with the look of things. While reading the article I thought it would be really cool it more Hollywood production companies were to get involved in this sort of academic work. Favro explained how often viewers in the audience would propose the idea or request that their be other walking figures in the virtual maps or portals, and while the graphics students would try over and over it never quite looked right and would distract from the buildings and narration. Favro states that “currently, there are no accepted scholarly guidelines for interaction with digital reconstructions,” I believe that a set or limited vocabulary and some guidelines could be very helpful so that a standard is created for what is academic reconstruction of a scene and what is fantasy based off historical happenings.

Favro claims that knowledge and understanding is forged and strengthened through movement” and that the “study of ancient sculptures and inscriptions must consider observes kinetic, experimentally rich engagement within the original physical context.” I agree with completely because as a artist I know that the way you show something is almost as important as what you are showing. Placement, height and everything like that has a huge effect on how people interpret what is being exhibited. For example, if an artist scatters all the paintings on the floor in the center of the room or if they hang them all level in elegant frames, the audience will perceive them differently and walk away with a different story to tell. If someone were to make a 3D model of a statue, it is important to place the viewers eyes in relation to where a typical person’s eyes might be. It would be really cool if they could add a feature where you could type in your height and then see if from that perspective, as if you were actually there. The UCLA students and faculty who worked on the UCLA project wanted to make cool bird’s eye views’ of the locations, but it seemed to break the continuity of the “walk through” idea that had originally come up with. What I found like it is the tour of the Louvre, which was really popular when it came out and you can see how things look in the context of the building. My favorite part about digitizing this sort of work is that now people who might be unable to fly to France (say the sick, elderly or disabled) can still see the work and better imagine how it might feel to be there  🙂

ALSO My computers having a heart attack so there will be no images for this post
If you want to see cool things, pelase look here:



Artist, dancer, stand up comic and improviser :)