I just wrote a several-hundred-word blog post explaining the hows and whys of the nothing-happeningness of it all, but I can't post because the people I ridicule in it could still make our lives even more difficult if they felt like it. So you don't get to know. Smiles, everyone smiles!
We tore down some drywall and found this buried underneath it. I wonder how many years it has been hidden in there.
It starts going dog-shaped at around 1:45, and we're fully floating in a cosmic background radiation froth of doglizards by 3:45.
I lied. It wasn't tricky at all.
Flynn and co's new approach is to train a machine vision algorithm to work out what the new image should look like having been trained on a vast dataset of sequential images.
They trained their algorithm using "images of street scenes captured by a moving vehicle." Indeed, they use 100,000 of these sequences as a training data set.
They then tested it by removing one frame from a sequence of Street View images and asking it to reproduce it by looking only at the other images in the sequence. Finally, they compare the synthesized image with the one that was removed, giving them a kind of gold standard to contrast it with.
The results are impressive. "Overall, our model produces plausible outputs that are difficult to immediately distinguish from the original imagery," say Flynn and co.
It comes with a $75/month subscription fee, because their actual business model is "capture your customers' personal information so that they can turn your customers into their product."
Remember when companies used to try to make money by selling you things, instead of by selling you out?
