JPEG Workshop on Subjective Quality Assessment Announcement
Thanks to the increasing storage capabilities and internet speed, modern image codecs have made very high-quality images possible and desirable. However, the existing standardized assessment methodologies, e.g. ITU-Rec. BT.500, are more suitable for evaluating the visual appeal of images (how obvious and/or annoying the artifacts are) than for evaluating their visual fidelity (how true the images are to the original). The methodology described in AIC-2, on the other hand, approaches the problem from the other end: it is based on a very sensitive flicker test that will catch even the slightest visual distortion.
In this context, the JPEG Committee has launched a new activity on Assessment of Image Coding, also referred to as JPEG AIC. This activity is a continuation of the previous standardization efforts (AIC-1 and AIC-2) and aims to develop a new standard, known as AIC-3. The scope of AIC activities is to specify standards or best practices w.r.t. subjective and objective image quality assessment methodologies that target a range from high quality to near-visually lossless quality.
- Date: June 28, 2022
- Time: from 15:00 to 17:00 UTC
- Access: Zoom
- 15:00-15:10 UTC - Opening (Evgeniy Upenik, Huawei)
- 15:10-15:25 UTC - JPEG AIC-3 Use Cases and Requirements (Jon Sneyers, Cloudinary)
- 15:25-15:40 UTC - Review of Subjective Image Quality Assessment Methods (Michela Testolina, EPFL)
- 15:40-16:10 UTC - Subjective Image Quality Assessment with Boosted Triplet Comparisons (Dietmar Saupe, University of Konstanz)
- 16:10-16:25 UTC - Unreasonable Effectiveness of Deep Features as a Perceptual Metric (Richard Zhang, Adobe)
- 16:25-16:40 UTC - MPEG VQA Remote Experts Viewing (Mathias Wien, RWTH Aachen University)
- 16:40-17:00 UTC - Q&A Panel
Interested parties are invited to join the JPEG AIC mailing list and regularly consult the JPEG.org website for the latest news.