Google is unofficially working on a new lossy image format for the internet (of course), called Pik. According to the ReadMe on GitHub, the directory contains an encoder and a decoder for the format. The company does warn that the project is in the initial research stage and that it shouldn’t be used for any purpose.
The ReadMe file goes on to explain that Pik requires an AVX2 and FMA capable processor, such as Haswell. Despite Google’s warning not to use the software, it is actually usable if you know how to build the source code, instructions are given for anyone wanting to get their hands dirty (and images totally wrecked!).
For those not familiar with the terminology, a lossy image format, according to Wikipedia, is:
“[A] class of data encoding methods that uses inexact approximation and partial data discarding to represent the content. These techniques are used to reduce data size for storage, handling and transmitting content … Well-designed lossy compression technology often reduces files sizes significantly before degradation is noticed by the end-user.”
Google’s current lossy (and lossless) image format, WebP, was announced in the September of 2010. It’s not clear whether Google hopes to deprecate WebP in favour of Pik at this point in time.