Workshop days are a bit full-on, so I don’t seem to have so much time to write the blog. I seem to be a bit drained by the evening! Fortunately we’ve got some nice videos which I think somewhat speak for themselves, but here are a few memories, thoughts, observations:

We’ve now reached the end of two workshops with professional choreographer / dancers. These proved extremely useful in highlighting issues (strengths and weaknesses) with me and my shadow as it stands so far, and pointed up many ideas for future development.  Thanks, boDig and MADE, for making these workshops and this residency possible.  It’s been great to develop the project with so much input, thought and enthusiasm from others.

Yesterday we spent quite a long time setting up the space (quite tricky- the Kinect can be a fiddly little blighter).  We then spent most of the session going through some quite rigorous exercises with Ghislaine Boddington.  These are exercises, or games (because they were fun too!) that she’s developed primarily in telepresence projects, some of them with myself.  They really help in getting used to working with and through the camera, and the relationships with space, screen and others the situation throws up.  In this instance they highlighted both the similarities and the differences between what we’re doing and video-based telepresence.  The crucial one is that the ‘real’ camera (the kinect) and the virtual one are totally independent, so the viewpoint shown by the video can be anywhere, entirely at odds with the physical placement of the Kinect.  This is both extremely exciting and rather challenging.

You’ll see from the videos below that we kept the virtual camera pretty much static for the first workshop.  Today, having established a strong orientation with real and virtual spaces, we were able to free things up a little bit.  Here are some of the things we explored:

1) we turned off the ‘tracers’ (the particles that trail behind the skeleton points) and just focused on the actual representation of the body in the system.  We found that it’s extremely different depending on distance – close to the camera, the body looks quite solid, and really quite detailed – you can actually recognise someone, and facial details, clothes etc. are quite deliniated.  As you get further away you become much more abstract, and the inaccuracies of the Kinect much more pronounced.  Yes, it’s obvious, but we found some very interesting results juxtaposing near and far, and playing with the rather distorted depth of field of the Kinect camera.

2) we tried out various combinations of tracers.  Overnight, Phill programmed in the capability to turn them on and off.  We found that less is definitely more, and the most interesting point we found to track was the central point of the spine.  This was really interesting to discover.  I now have an embrionic idea that we should represent the various tracers differently – some more prominently than others.  I like the idea that there may be a ‘trunk’ tracer (the spine, say), with little filaments branching off it to represent the others.

3) It’s really interesting the way people and things appear and disappear. The Kinect is surprisingly fussy about this.  As you edge into frame you won’t appear until there’s enough of you visible for the Kinect to recognise as a human form, at which point you’ll suddenly pop into existence.  The reverse can happen too, and relationships between people and objects can do strange things – kind of turning each other off and on.  It can be an interesting phenomenon, and also does strange and rather satisfying things to the sound (yes, we have sound now, although it needs a lot of refining) as lots of points of sound appear or disappear at once.

4) the virtual camera is both the most challenging and the most interesting thing.  It’s fascinating to look at the body and physical movement from unexpected viewpoints, but it can also be very confusing.  Over the course of today, I felt that Phill developed quite a skill as a ‘virtual cameraman’, choreographing the movement of the virtual camera expertly and artistically with the dancers.  This makes things look much more interesting – the videos above (today’s) look far more dynamic and three dimensional than yesterdays when the camera was largely static.

Many thanks to the dancers – Beliz, Banu, Dilek and Steven.  Also to Ghislaine, Phillipe, Phill and Yigit.  You’ve all made these couple of days fun, rewarding and extremely worthwhile.