CAE-ADMM: Implicit Bitrate Optimization via ADMM-based Pruning in Compressive Autoencoders
This work addresses image compression efficiency for applications requiring high-quality visual data, but it appears incremental as it builds on existing compressive autoencoders with a novel optimization approach.
The paper tackled the trade-off between distortion and efficiency in lossy image compression by introducing CAE-ADMM, which uses ADMM-based pruning to implicitly optimize bitrate, and it outperformed the original CAE and some traditional codecs in SSIM/MS-SSIM metrics.
We introduce ADMM-pruned Compressive AutoEncoder (CAE-ADMM) that uses Alternative Direction Method of Multipliers (ADMM) to optimize the trade-off between distortion and efficiency of lossy image compression. Specifically, ADMM in our method is to promote sparsity to implicitly optimize the bitrate, different from entropy estimators used in the previous research. The experiments on public datasets show that our method outperforms the original CAE and some traditional codecs in terms of SSIM/MS-SSIM metrics, at reasonable inference speed.