Skip to main content


 
Both interesting and concerning to see the advancement of technology and techniques that can provide greater accessibility yet also fake lip sync that is so difficult to detect as fake.

OMFG! That's such a good #deepfake ... i love it!

Canny AI: Imagine world leaders singing


#video #music #johnlennon #imagine #deeplearning #ai #art

It's not a comprehensive idea but the videos used have the same main color profile because only the lips are adjusted. A search could match to the original content. Simply cropping might thwart that though. I guess extensive watermarking signatures for peoples original content might do the trick.

BTW: Just saw this piece (surprisingly from an eBank) about the #Deepfakes problems and half-way down the article they discuss two active measures currently combatting the problems.



 
verryyyy interesting ;)

#AI #surveillance #escape

This colorful printed patch makes you pretty much invisible to AI - The Verge


... show more

@Shelenn Ayres I have so experienced that...riding on my bike...late at night

When the White Cop stopped me, it was over in 5. Unlike the normal "swarming" of police, in this college town.

It's like the image he's hanging around his neck made him a building relative to the indiscernible pastel people in the image. I suspect a whole new genre of clothing will be coming out soon.



 

AI for Fun: Awesome Web and Mobile Apps You Can Test Right Now


Artificial Intelligence and Machine Learning. What is it? Just another hype, modern trend to name the technology, or maybe, it is a potential superpower to kill humans?! Today we hear those buzzwords and what is more serious, wrong assumptions about them from everywhere, but only a few understand what it really is and how it works. And if
... show more



 


GauGAN turns your doodles into photorealistic landscapes

NVIDIA's deep learning AI model GauGAN, cleverly named after post-impressionist painter Paul Gauguin, turns simple sketches into realistic scenes in seconds by leveraging generative adversarial networks, or GANs, to convert segmentation maps into lifelike images.

GauGAN allows users to draw their own segmentation maps and manipulate the scene, labeling each segment with labels like sand, sky, sea or snow. The tool also allows users to add a style filter, changing a generated image to adapt the style of a particular painter, or change a daytime scene to sunset.

Source: https://blogs.nvidia.com/blog/2019/03/18/gaugan-photorealistic-landscapes-nvidia-research/

#AI #deep learning #art

What's it do when you try and draw a face, though?

lol good question!



 
Artificial AI 'Fingerprints' Can Fool Biometric Scanners

Despite their popularity, and a wide perception of offering strong and reliable security, biometric security technologies which measure things like fingerprints, faces, eyes, etc... are not without their vulnerabilities.

One arg
... show more