Projects
Projects
Embedded Files
DeepCQ: A Two-Stage Deep-Surrogate Framework for Lossy Compression Quality Prediction
Khondoker Mirazul Mumenin, Robert Underwood, Dong Dai, Jinzhen Wang, Sheng Di, Zarija Lukic, and Franck Cappello
Cluster, 2026
we present a modular two-stage deep-surrogate framework that produces accurate compression quality prediction across widely used lossy compressors and quality metrics, enabling users to bypass costly compression operations during configuration selection.
Github Repo, Talk
QualityNet: Error-bounded Lossy Compression Quality Prediction via Deep Surrogate
Khondoker Mirazul Mumenin*, Dong Dai, Jinzhen Wang, Sheng Di
BigData, 2024
We designed a surrogate-based framework to predict key compression quality assessment metrics. This framework allows users to estimate the quality of compressed data quickly, which enables faster decision-making on compression configuration.
Github Repo, Talk
Page updated
Report abuse