- Check the submission instructions and reporting guidelines.
- Registration is now open ! Register your proposal here
- Check the new timeline and the awards.
- JPEG AI Call for Evidence is available
- Website up and running
The main objective of this challenge (and JPEG Call for Evidence) is to objectively and subjectively evaluate relevant learning-based image coding solutions to demonstrate the potential of this coding approach, especially in terms of compression efficiency. This topic has received many contributions in recent years and is considered critical for the future of image coding. Naturally, improvements on some aspects (mainly tools) of existing learning-based image codecs are also welcome.
Participants are required to submit the following material:
- A detailed description of the coding algorithm in the form of an MMSP paper and/or JPEG technical report as well as methodologies and data used for training.
- A decoder implementation in a form that allows stand-alone inference/testing on a standard computer (CPU only) in a reasonable amount of time, preferably in source code form.
- Compressed codestreams and corresponding decoded images.
The MMSP 2020 submissions will be peer reviewed in a similar way as any other regular MMSP paper submissions and, if accepted, will go to IEEExplore.