Google is developing a new WebP 2 image format

Google has published the works related to a new experimental image encoding format called "WebP 2", which is developing as a more efficient replacement for the WebP format.

Since the new format is still in development and has not been finally defined, so not yet ready for widespread use (Backward compatibility on encoder and decoder is not guaranteed, the code has not been optimized.)

About WebP 2

En WebP 2 describe new features for your implementation, such as HDR with 10-bit color representation, more efficient compression of transparency information, full support for animation, easy incremental decoding (layer-by-layer decoding with greater detail at each stage, allowing you to generate thumbnails very quickly for preview), fast multi-threaded software implementation, minimization of visual degradation at low rates of bits, improved lossless compression mode.

WebP 2 is the successor to the WebP image format, currently in development. It is not ready for general use and the format is not finalized, so changes to the library may break support for backward-encoded images. 

This package contains the library that can be used in other programs to encode or decode Webp 2 images, as well as command line tools.

The purpose of the new format is identical to that of the first WebP: transmission of images over the network, optimization for medium resolutions, use in web and mobile applications, with support for common tasks for these tasks, such as support for transparency, animation and quick sketches.

The experimental WebP 2 codec primarily drives WebP's features in terms of compression efficiency. New features (such as 10b HDR support) are kept to a minimum. The axes of experimentation are:

More efficient lossy compression (~ 30% better than WebP, as close to AVIF as possible)
Better visual degradation at very low bitrate
Improved lossless compression
Improved transparency compression
Animation support
Ultralight previews
Light incremental decoding
Small top container, specifically designed for Image Compression
Full 10-bit architecture (HDR10)
Strong focus on software implementation, fully multithreaded
The use cases remain mostly the same as WebP: wired transfer, faster web, smaller applications, better user experience… WebP 2 mainly fits typical content available on the web and mobile applications: range dimensions medium, transparency, short animations, thumbnails.

The main efforts in the development of the new format aim to increase compression efficiency.

First WebP achieves file size reduction from 25% to 34% compared to JPEG files of similar quality, and in lossless compression mode, it achieves a 26% reduction in the resulting file size compared to the maximum compression level of PNG. WebP 2 aims to achieve an improvement efficiency 30% lossless compression compared to the first WebP and bring the AVIF lossy compression codec to 20%.

The prototype under test is still poorly optimized and it falls far behind the polished implementation of libwebp in terms of encoding and decoding speed. For example, in lossy compression mode, WebP 2 compresses five times slower than the first WebP.

Compared to libavif, the new WebP format what is developing Google encodes twice as fast, but it lags 3 times in decoding speed. At the same time, by the time the final version of the libwebp2 library is released, it is planned to achieve parity in decoding speed.

Finally, for those who are interested in knowing more about the note, they can consult the original publication In the following link.

And as for those who are interested in knowing the project code, as well as its progress, they can consult it by going tol following link.


Leave a Comment

Your email address will not be published. Required fields are marked with *

*

*

  1. Responsible for the data: AB Internet Networks 2008 SL
  2. Purpose of the data: Control SPAM, comment management.
  3. Legitimation: Your consent
  4. Communication of the data: The data will not be communicated to third parties except by legal obligation.
  5. Data storage: Database hosted by Occentus Networks (EU)
  6. Rights: At any time you can limit, recover and delete your information.