Enhancing Perception Quality in Remote Sensing Image Compression via Invertible Neural Network

TL;DR

This work proposes a novel invertible neural network-based remote sensing image compression method that captures the compression distortion from an existing image compression algorithm and encodes it as Gaussian-distributed latent variables using an INN, ensuring that the distortion in the decoded image remains independent of the ground truth.

Abstract

Decoding remote sensing images to achieve high perceptual quality, particularly at low bitrates, remains a significant challenge. To address this problem, we propose the invertible neural network-based remote sensing image compression (INN-RSIC) method. Specifically, we capture compression distortion from an existing image compression algorithm and encode it as a set of Gaussian-distributed latent variables via INN. This ensures that the compression distortion in the decoded image becomes independent of the ground truth. Therefore, by leveraging the inverse mapping of INN, we can input the decoded image along with a set of randomly resampled Gaussian distributed variables into the inverse network, effectively generating enhanced images with better perception quality. To effectively learn compression distortion, channel expansion, Haar transformation, and invertible blocks are employed to construct the INN. Additionally, we introduce a quantization module (QM) to mitigate the impact of format conversion, thus enhancing the framework's generalization and improving the perceptual quality of enhanced images. Extensive experiments demonstrate that our INN-RSIC significantly outperforms the existing state-of-the-art traditional and deep learning-based image compression methods in terms of perception quality.

Figures (10)

• Figure 1: Overview of existing INN-based image compression methods. For simplicity, entropy models are omitted. It can be seen that our method is a plug-and-play (PnP) approach that does not require retraining of existing compression models. Our method can obtain both decoded images with high PSNR and MS-SSIM output from the original compression algorithm (e.g., ELIC) and enhanced images with high perceptual quality.
• Figure 2: Visualization of the decoded images by ELIC he2022elic and the proposed INN-RSIC.
• Figure 3: The architecture of the proposed INN-RSIC. Compressor refers to the competitive image compression algorithm he2022elic.
• Figure 4: Illustration of the conditional generation module (CGM).
• Figure 5: Performance comparison on the testing set of DOTA and UC-M. It is observed that INN-RSIC achieves a better trade-off between fidelity (i.e., higher PSNR) and perceptual quality compared to HiFiC.