SIC (Structured Image Compressor) lossy codec

SIC VERSION 0.201

Welcome to the homepage of SIC (Structured Image Compression), an innovative image codec project focused on re-evaluating priorities in digital image compression. Developed with a strong emphasis on computational efficiency during

encoding and high-quality reconstruction at aggressive bitrates, SIC explores distinctive architectural approaches compared to conventional state-of-the-art solutions. His new version (x64 exe) is visually structured differently from the previous one in that it is aimed at better image stability in every possible context. I used a fixed primary set of 44 variegated images to essay the improvements of the various experimental versions. Both compressine and decompression speed increased. The version under test is compiled with GCC 15.2. I am waiting to assay your impressions.

Core Features of the SIC Codec

The SIC codec leverages a sophisticated approach to image compression, primarily built around a highly optimized OGBT Transform . Unlike many conventional codecs, SIC dynamically applies OGBT blocks  16x16 size, for optimal processing. Currently, SIC's design is focused on single-image compression, and does not yet incorporate advanced intra-frame prediction or inter-frame prediction (for video sequences). Similarly, a deblocking filter is not currently implemented. However, future support for video encoding is not excluded.

The paramount strength of SIC lies in its exceptional compression efficiency, achieved through an advanced entropy coding engine featuring aggressive range coding. This powerful mechanism allows for highly efficient compression of OGBT "Orthogonalized Adaptive Gabor Basis for Energy Compaction in Two-Dimensional Block Transform" coefficients, resulting in remarkably small file sizes while maintaining competitive image quality. In many test cases, SIC has demonstrated superior performance compared to the established WebP format, and its results consistently remain highly competitive, not significantly deviating from the advanced AVIF codec. I added full support for 32-bit images (ARGB/BGRA) thus including Alpha channel images characterized by transparencies. 

SIC remains an active research and development project. Future efforts will focus on exploring further optimizations within its unique architectural framework, aiming to push the boundaries of computational efficiency while maintaining and enhancing its already strong quality profile.

Multithreading support . From version 1.0 I will be able to guarantee backward compatibility.

IMPORTANT Apply: State of development
Note: Sic is currently in an experimental and active development phase. As such:

The backward compatibility is not guaranteed before the official release 1.0. File formats and API interfaces can change.

We do not recommend using SIC to codify images of personal or professional critical interest in which long -term conservation or universal compatibility is required. This codec is mainly intended for specific research, tests and applications in which its unique strengths are beneficial and the above limitations are included.

 Home  page of SIC

Forum:   

Encode.su  

Reddit

 Doom9 

If you are interested please contact me at the email to : 

  miadebora@hotmail.it