Synthesised reality

[carllooper]

While not directly related to small gauge filmmaking I wanted to share this anyway.

It's part of research I've been doing over many years in the realm of image processing, as those on this forum might know. In particular I've been very much into how Super8 (or any other gauge) might be processed in the digital domain. The following is one of the tangents I've been taking in such research. It's not Super8 but it's part of research towards algorithms that would operate on Super8.

The main thing here is that the videos are not computer generated in the usual sense. Rather, they originate in natural photographs (in this case digital photographs) rather than in some 3D artist's imagination. That's an important distinction. Photography proper, in my books, is about images that originate outside of our imagination. Not in some notion of objective reality (I reject that as well) - just simply outside of subjectivity.

The results here are basically the result of "stitching" together 20 individual photographs of the same tree, each taken at different angles. The software estimates, from each photograph, what the three dimensional shape of the tree would be (quasi-analogous to the way our brain might do the same, but I have no idea whatsoever how our brains actually work). After that, the individual images (photographs) are then projected back onto the three dimensional shape, to form a single volumetric image in virtual space. I then create a new camera angle on such - for example, one that might move around the tree, and render such:



Video: https://vimeo.com/150200961



Video: https://vimeo.com/150231845

[slashmaster]

So ultimately, you're talking about filming an area with super 8, stitching it all together, then visiting it with the Oculus Rift or some other virtual reality headset? Now that you bring this subject up, I'm very excited about virtual reality but afraid it's going to be a major blow to film

[carllooper]

So ultimately, you're talking about filming an area with super 8, stitching it all together, then visiting it with the Oculus Rift or some other virtual reality headset? Now that you bring this subject up, I'm very excited about virtual reality but afraid it's going to be a major blow to film

I don't see it as an either/or situation. New technology and old technology can operate at the same time. The invention of photography didn't kill painting. And yet painters of the day thought it would do just that. Looking at a photograph they fully believed that was the end of painting. But it was only the end of a certain type of painting. Painting recovered and went on to explore new ways of working with paint that would distinguish itself from photography.

And photography, for it's part, stopped trying to pretend it was a substitute for painting and found what it was actually good at.

New technology is disruptive but it doesn't always kill off a previous technology. It can just change how older technology might be otherwise used. Its up to filmmakers working with film to find new ways of working with film and to challenge new technology, such as digital, to pursue it's own particular path.

Re. Super8. That's a good idea - shoot a scene on Super8 from a thousand different angles and combine it all together into an Occulus Rift experience. I've written software for a couple of Occulus Rift jobs and it's kinda interesting. The screens on the Rift are pretty crappy but the potential is certainly there.

Here though it's a bit of a different agenda. I'm more interested in the temporal aspects of film - more than it's spatial aspects. Or to put it another I'm more than happy with the 2D space of film. However I do explore the n dimensional aspect of space (as is obvious), but only as a precursor to it's eventual re-rendering back into the 2D space of film. And then reframed in terms of time (music, narrative, etc). Basically I like the cinema as a venue for work. Not that these two examples actually demo any of that. They are just tests - to see how the algorithms are working - and what needs more attention etc. But I like them as works in their own right as well.

C

[Simon Lucas]

Carl, these look beautiful. They have a lovely quality – a kind of dream-like ambivalence. I look forward to seeing where you go next with this idea.