Will Thomas Israel go really generative?
(A light revision in Jan.2017)
Israel : At the Woods Border
From september 13 to october 13, Thomas Israel will hold a solo exposition at Galerie Charlot (Paris), entitled "A la lisière du bois". That let's us expect that his work Lisiere will get the focus, keeping in line with the general orientation of this Gallery onto combinations of video and physical sculpture (like the works of Hugo Verlinde or Misha Margolis, for instance).
But perhaps will he give also some views of the generative track he opened with his Méta-Crâne work (First show 2009), which is the most in line with my present reflections. In the intricate mesh of themes and techniques that Israel weaves along the years, he tied a new node, of impressive potential. His whole videographic production during the last five years (2004-2009) has been cut into 1 000 films, 100 musics and 80 lyrics. The films have been evaluated according to eight criteria : hue, saturation, value, internal speed (defined by the amount of pixel change from frame to frame) and the fluctuation of these four inside of each media Then, a specially designed software creates an indefinite length of video projection as a succession of the clips according to some laws (for instance, a saturated clip will be followed by a more saturated clip).
Israel : Metacrane
The webtsite do not tell how musics and lyrics are associated with the video clips. We have here a typical example of art generation from a set of parts, following what Bootz calls "combinatorial" (combining pre-built meaningful fragments such as sentences), as opposed to "automatic" (combining basic elements such as letters). These methods have been mainly used for text (fo instance in France by Oulipo) and for images, be it in a rather coarse and random way (Cadavre esquis) or more "consciously" (collages, matte painting).
As far as we know, Israel is the first to do that for video, associating sound and image. From a generative standpoint, the fact that Israel embeds this formal core in a rather costly contraption (customized cabin for 360 degrees projection and 5.1 surround sound system) and presents it with a semi-scientifical speech ("free association", "metaphor of the mind", "representation of the artists real brain"...). The really important basis here is the algorithmics (the grammar, somehow) of clip succession and sound association, plus the sound environment.
The analysis and classification of the medias is done by a special pluggin created for Meta-crane by Numédiart engineer Xavier Siebert, depending of their Media-cycle project. The interactive edditing of sound and vidéo, linked to the activity of the visitor in the skull, have been written by the artist on Isadora. We think it could go much farther with less hardware and speech and more software. As fo hardware, it would not lose much (and possibly gain up) if presented on a mere screen with a sound headset, allowing to reach wider publics (including home cinema and even cheaper installations).
As for software (logic and contents) the field of exploration and artistic creation/enjoyment could enlarge out with :
- a larger set of fragments (clips, musicals, lyrics), possibly adding clips other than the past films of the author, and why not combined with real time recording at the same time,
- a wider range (but we have no data on this point) of transitions (cut, crossfade, inlay...),
- a deeper call to interaction, which remains limited in Métacrâne, even if it follows quite well the recommendations of Aziosmanoff, for instance,
- a richer space of criteria (I think of course of the coarse complexity criterium I use in Roxame),
- and more than else (but it is not easy) a call to semantics (pattern recognition with feature space, dictionary...) to put meaning into the compositing game.
He does not lends himself to easy prediction, since we see him working with glue or a solder iron as well as performing live on stage. He could be compared to Lozano-Hemmer or Matt Pyke, with their large choice of technologies applied to persevering thematics (surveillance for Lozano-Hemmer, rythmics fof Pike), constrasted with his psychological, sex-self-conscious and rather dark ways. Anyway, they all add to their own skills through cooperation with other specialists (music composers, DJs, professional programmers). And the evolution of their oeuvre is certainly partly driven by their fans and customers demands. Thet let us guess that "free association" will remain a theme among others in his flow of creativity. For the time being, it will be interesting at Galerie Charlot to get a look to his "Lisière" work and to whatever he will show and perhaps say about.
9/13/2012
DICCAN'S PARTNERS:
Paris ACM Siggraph, the French chapter of ACM Siggraph, worldwide non-profit organization of computer graphics.
Les Algoristes, an association of artists using their own algorithms in their work.
