At Flipboard, we did not have time to do a full comparison of related upscaling research, but we were happy with the low amount of error our CNN achieved.
FWIW, waifu2x was inspired by this paper from researchers at Chinese University of Hong Kong and Microsoft Research Asia http://arxiv.org/abs/1501.00092v3
waifu2x is a great demonstration of this approach applied to a specific domain.
By coincidence, Flipboard's DNN approach was developed around the same time as the MSRA research in summer 2014.
I'm excited to see future research in applying deep learning to generative tasks. Some of the CNN music composition work is quite impressive.
Back at Flipboard, for our RSS typography processing, we used a rule of thumb that all-capped words 3 letters or longer would get small-caps treatment. It worked pretty well.
Generally, but not in all cases. IANAL, but if you are an officer or key employee of a company that's acquired, the acquirer can enforce non-competes because it affects the value of the acquisition. There's a specific section of the California employment code that talks about this, but I don't have it handy right now.
Understood. We just happened to have unconventional constraints that suggested an unconventional solution. There are many ways to lobby for improvements for things (especially Web standards) and perhaps with us showing the limitations of the DOM, we can influence change for the better.
On iOS, updates to the UI happen on the main thread, but all animation, surface compositing, and rendered graphics run on the dedicated Core Animation rendering thread.
We're seeking iOS and web developers with a passion for design and craftsmanship. We're doing crazy things at the intersection of native and web platforms in the name of user experience. If that appeals to you, let me know!
We're seeking iOS and web developers with a passion for design and craftsmanship. We're doing crazy things at the intersection of native and web platforms, and if that appeals to you, let me know!