Pik
Pik is a new lossy image format for the internet. This directory contains an encoder and a decoder for the format.
This project is in the initial research stage, please don't use it for any purpose.
Project Goals
Pik is to have roughly the same performance requirements space that JPEG is holding today. We aim to have roughly similar decoding speed, i.e., no more than 50 % slower in decoding, but adding more modern compression and prediction methods, giving 55 % more density. We plan to improve to 65 % before freezing the format. Some of these new advances may come with a moderate drop in decoding speed (possibly 40 % of jpeg speed).
We are planning for a tiled format so that multi-core decoding can decode a single image faster (possibly around 16x). The current version has no support for this yet.
We are planning to keep the format 8x8 DCT based, possibly with some support for non-integral-transform-based direct mode blocks (or overlay blocks).
Build instructions
The software currently requires an AVX2 and FMA capable CPU, e.g. Haswell.
Please ensure you have the libpng-dev and libjpeg-dev packages installed.
Then simply run make -j8, which creates cpik and dpik binaries in bin/.
The second command line argument to cpik is a Butteraugli distance (see http://github.com/google/butteraugli), which indicates the largest acceptable error. Larger values lead to smaller files and lower quality. Try 1.0 for a visually lossless result.
Related projects
- Butteraugli (HVS-aware image differences)
- Guetzli (JPEG encoder with denser packing)
This is not an official Google product.