note it also turns the pasture into savannah - another great example of dataset sample bias influencing an ML model.pic.twitter.com/VVttFLF8lB
U tweetove putem weba ili aplikacija drugih proizvođača možete dodati podatke o lokaciji, kao što su grad ili točna lokacija. Povijest lokacija tweetova uvijek možete izbrisati. Saznajte više
note it also turns the pasture into savannah - another great example of dataset sample bias influencing an ML model.pic.twitter.com/VVttFLF8lB
This is also because it's totally unsupervised so it doesn't even know the horse is the only part we want it to care about
That is one crazy zebra tail… Seriously though, I am astounded at that.
is this real life? is this fantasy! This is so cool
I'm surprised at the temporal coherency, it looks much better than style transfer on videos did. But wasn't it trained on images, not video?
Ok. That's truly badass! (Horse changed to Zebra in video) https://github.com/junyanz/CycleGAN … @AWSOMEDEVSIGNER @benriga @stephenlb @girlie_mac
Very cool. A little weird to see the stripes on the tail. Crazy that they can do this on video.
I'm really surprised by the time coherency of stripes
Woo! but I'm even more impressed by the Monet to photo stuff
https://camo.githubusercontent.com/2296236e17ff15eb5a077fdb62df498b9d000a19/68747470733a2f2f6a756e79616e7a2e6769746875622e696f2f4379636c6547414e2f696d616765732f7061696e74696e673270686f746f2e6a7067 …
Do a unicorn. Do a unicorn 
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.