Simon Niklaus, Long Mai, Jimei Yang, and Feng Liu developed software that gives a 3D Ken Burns effect to still images.
From their paper:
Experiments with a wide variety of image content show that our method enables realistic synthesis results. Our study demonstrates that our system allows users to achieve better results while requiring little effort compared to existing solutions for the 3D Ken Burns effect creation.
Denis Shiryaev writes that this footage, filmed in Tokyo between 1913-1915 as “Japan of Today”, was upscaled using neural networks. ✔ FPS boosted to 60 frames per second, I have also fixed some playback speed issues; ✔ Faces are enhanced too – I have added to the pipeline of algorithms a neural network which is […]
How does a neural network connect perceptions to concepts? In other words, how can you make something that accepts an array of pixels as an input and correctly outputs “dog” or “cat?” This video from Art of the Problem does a good job of explaining how neural networks are able to do this, and why […]
DICK-RNN is a recurrent neural network trained with 10,000 doodles of dicks. You start drawing a shape and the AI tries its best to finish it as a penis. Starting with the balls helps a lot. Try it. For the technical background and conceptual inspiration: GitHub page for DICK-RNN
Even after months of working from home, you’d be forgiven for thinking the whole experience still doesn’t quite feel…well, normal. In addition to all the obvious environmental changes of handling your 9 to 5 from your den or dining room table, the technological aids you didn’t realize you loved back at the office probably don’t […]
Running a small business drops a lot on to the plate of just one person. And between juggling a dozen tasks that need to get handled daily, it’s no surprise that there are a dozen more equally vital tasks that can just as easily go overlooked. While posting to social channels and making web posts […]
The importance of reading is well documented. About half of America’s unemployed between 16 and 21 years old are functionally illiterate. And there’s an almost direct line between how much you read and your earning potential, with the richest Americans three times more likely to read than those with a household income below $30,000. However, […]