Google Replicates “Silicon Valley’s” Image Compression Algorithm

Google have taken on the likes of HBO’s Silicon Valley technology this week with its introduction of an AI based image compression system that was seen on the show. Researchers have since released a paper entitled, “Full Resolution Image Compression with Recurrent Neural Networks.” In Silicon Valley, the system is called the Pied Piper by its creator, Richard Hendricks and can turn files into featherweight which changes the world of storage and file sharing. In the show, Hendricks makes it big because if this breakthrough.




Currently, compression methods are generally inefficient and lose information from the images while pixels degenerate often making the image appear faded or washed out. However, with Google’s AI image compression, full resolution is maintained. Google researchers comment, “As far as we know, this is the first neural network architecture that is able to outperform JPEG at image compression.”

The way in which it works is by having a neural net pass over the photograph several times. As it does this is takes in more information pertaining to the image and how to best downsize it. Google writes, “While today’s commonly used codecs perform well, our work shows that using neural networks to compress images results in a compression scheme with higher quality and smaller file sizes.”

So, although it might not be quite the same as featured in Silicon Valley, at least Google’s system is real, tangible and here now. Whereas the Pied Piper system is still merely a pipe dream.





More News To Read

Comments

comments