We at Cloudinary are delighted to share several exciting developments in image quality assessment and encoding, featuring our very own Head of Image Research Jon Sneyers and collaborative work across the industry.
Recently, a comprehensive paper titled “The JPEG XL Image Coding System: History, Features, Coding Tools, Design Rationale, and Future” was published, offering deep insights into the design and potential of JPEG XL.
You can explore this detailed resource here: JPEG XL Paper.
Our team published an insightful article in collaboration with researchers at the University of Konstanz, titled “AIC-3 and HDR image quality assessment.” The results from JPEG AIC-3 — JPEG’s methodology for fine-grained subjective quality assessments — highlight JPEG XL’s strengths and affirm SSIMULACRA 2 as one of the top-performing objective metrics for estimating subjective image quality.
Read more about it here: AIC-3 and HDR.
JPEG is continuing its innovation with the AIC-4 initiative, aiming to enhance objective quality metrics and establish a new standard. Interested parties are encouraged to participate in the ongoing call for proposals: AIC-4 Call for Proposals.
Additionally, the Alliance for Open Media recently released libaom version 3.12.0, now featuring a new quality tuning mode based on SSIMULACRA 2, acknowledging the metric’s importance in the industry.
Check it out here: libaom 3.12.0.
Jon Sneyers, the creator of SSIMULACRA 2 and one of the key developers of JPEG XL, continues to drive forward our efforts in image optimization. You can find more information about JPEG XL at the official website jpeg.org/jpegxl and at the community website jpegxl.info.
We are proud to contribute to this vibrant community and invite everyone to dive into these exciting advancements!
.pdf){kind=link}