CIC: Circular Image Compression

TL;DR

This paper introduces Circular Image Compression (CIC), a novel closed-loop approach that enhances learned image compression performance, especially on out-of-sample images, by reducing the gap between training and testing data.

Contribution

The paper proposes a closed-loop architecture for LIC, called CIC, which improves robustness and performance on diverse images, and can be integrated with existing SIC methods.

Findings

CIC outperforms eight state-of-the-art SIC algorithms in rate-distortion performance.

CIC effectively handles out-of-sample images with complex patterns and high contrast.

The method demonstrates near-zero steady-state error in image reconstruction.

Abstract

Learned image compression (LIC) is currently the cutting-edge method. However, the inherent difference between testing and training images of LIC results in performance degradation to some extent. Especially for out-of-sample, out-of-distribution, or out-of-domain testing images, the performance of LIC degrades significantly. Classical LIC is a serial image compression (SIC) approach that utilizes an open-loop architecture with serial encoding and decoding units. Nevertheless, according to the principles of automatic control systems, a closed-loop architecture holds the potential to improve the dynamic and static performance of LIC. Therefore, a circular image compression (CIC) approach with closed-loop encoding and decoding elements is proposed to minimize the gap between testing and training images and upgrade the capability of LIC. The proposed CIC establishes a nonlinear loop equation and proves that steady-state error between reconstructed and original images is close to zero by Taylor series expansion. The proposed CIC method possesses the property of Post-Training and Plug-and-Play which can be built on any existing advanced SIC methods. Experimental results including rate-distortion curves on five public image compression datasets demonstrate that the proposed CIC outperforms eight competing state-of-the-art open-source SIC algorithms in reconstruction capacity. Experimental results further show that the proposed method is suitable for out-of-sample testing images with dark backgrounds, sharp edges, high contrast, grid shapes, or complex patterns.