After the last post I was going to leave this for a day or so but I couldn’t help having a go. The natural next step for me, having gathered a bunch of photos together, was to pixelate them in some manner. I tried a basic 12×12 but it didn’t do much other than create a blocky smudge. So I pushed it further, resizing each photo as a 1×1 pixel square. In other words forcing the resizing algorithm to chose only one colour. Though I don’t have any idea how it decides I’m going to assume for the sake of this exercise that it’s the average colour by some definition of average. At the very least it’s an extremely abstract representation of the photograph.
On its own the square of colour means nothing. But when I put all these abstracted versions of the IMG_4228 photos together we finally see something. The video has a nice shimmer but it’s the canvas that finally tells us something. Gone are the aberrations, the inconsistencies in cropping and framing, the mud of sameness and the jarring of distinction. Nor does does the lizard brain get distracted by provocative naked flesh nor the protective urge triggered by cute child faces. Pattern matching is useless. The intellect has nothing to hook on to. No themes, no cues, no templates to project on.
After working with the chaos of the IMG_4228 slice I found this canvas calming. I’d been wondering if I might find anything at all in a sample which was by its very definition free of patterns. These photos have nothing in common with each other but in order to understand that I had to remove anything that might lead me to think they did. Now I can see the chaos.
I feel the urge to get scientific now. I want to take further samples from the IMG_XXXX spectrum and see how they compare to this. And then I’ll need to take samples where the title of the image implies a connection. What would average pixel canvas look like for capital cities like New York or London? How would the average pixel canvas for people’s names compare?
I think I’ve stumbled upon something here. And I’m quite excited.
This is great. I’ve no idea what it says about people, colour, technology, perception etc, but it ought to be saying stuff about all of these. I keep coming back to the thought that Flickr is ultimately “just” a photo sharing site. It’s not primarily a boasting board etc. That’s pretty obvious by the low artistic quality of much of what came up in your rake-through of it. That’s not a snobby remark from a photographer; I’m just making the point that people were posting what to the rest of us are pretty normal, random images, in the main, so as to share them with (presumably) a small target audience of friends or realtives. What that tells us about colour, I don’t know. But then, if people were posting a “best shot”, it might be easy (and probably wrong) of me to assume it would be more colourful, better composed, etc, with a consequent bearing on how it ultimately got rendered to a single pixel.
I’ve not the time or patience to do one of these myself, but if you feel you get the technique properly sorted, it would be fascinating to see what another one with a different selection produced, and muse on any differences.
There are the makings of a teriffic exhibition in this project!
It is a really interesting concept, makes for beautiful random images.
Have you been looking at the work of Jason Savalon? He did the the same thing with film stills, music videos and paintings, all of which are very beautiful too!
Looking forward to seeing more :D
Fascinating. I immediately wondered if there was a connecton between photos with similar emotional moods. That is, so they have a similar pixel square that our brains are subconsciously picking up while it processes the visual information?
Or maybe there is a connection pixel squares and visual punctum (is that the right term for the gut-related instant visual meaning that some photos offer whilst others leave the observer unaffected)? If so, you’ve hit on something rather wonderful!