State-of-the-Art Trends in Data Compression: COMPROMISE Case Study

Area of Science:

• Computer Science
• Information Theory

Background:

• Traditional data compression algorithms (e.g., JPEG, MP3) achieved high compression ratios but are now challenged by new computing paradigms.
• Emerging trends include digital data restoration and feature-based compression, requiring new approaches beyond domain-specific methods.

Purpose of the Study:

• To critically evaluate prominent new trends in data compression.
• To explore the parallels, complementarities, and differences among these trends.
• To introduce a novel methodology that addresses these challenges and integrates various approaches.

Main Methods:

• The study critically evaluates emerging data compression trends like digital data restoration and feature-based compression.
• A new methodology, COMPROMISE, is developed to integrate these trends and existing methods.
• COMPROMISE is designed to be domain-independent, asymmetric, and universal, supporting lossy, lossless, and near-lossless compression.

Main Results:

• Existing data compression methods face limitations with modern computing paradigms like cloud and edge computing.
• Digital data restoration and feature-based compression offer new avenues for data compression.
• A unified, domain-independent method addressing these trends is currently lacking.

Conclusions:

• The COMPROMISE methodology offers an interoperable framework for diverse data compression challenges.
• It integrates digital restoration and feature-based compression, promoting a general, universal approach.
• COMPROMISE aims to link existing methods, support hybrid techniques, and foster future algorithm development.