Bi-Level Spatial and Channel-aware Transformer for Learned Image Compression

TL;DR

This paper introduces a novel Transformer-based image compression method that incorporates frequency-aware attention mechanisms and a mixed feed-forward network, significantly improving compression efficiency over existing learned image codecs.

Contribution

The paper proposes a hybrid spatial-channel attention Transformer with frequency-aware modules and a mixed local-global network, advancing the transformation stage in learned image compression.

Findings

Outperforms state-of-the-art LIC methods in rate-distortion metrics.

Enhances feature decorrelation through frequency-aware attention mechanisms.

Improves compression efficiency with novel Transformer architecture.

Abstract

Recent advancements in learned image compression (LIC) methods have demonstrated superior performance over traditional hand-crafted codecs. These learning-based methods often employ convolutional neural networks (CNNs) or Transformer-based architectures. However, these nonlinear approaches frequently overlook the frequency characteristics of images, which limits their compression efficiency. To address this issue, we propose a novel Transformer-based image compression method that enhances the transformation stage by considering frequency components within the feature map. Our method integrates a novel Hybrid Spatial-Channel Attention Transformer Block (HSCATB), where a spatial-based branch independently handles high and low frequencies at the attention layer, and a Channel-aware Self-Attention (CaSA) module captures information across channels, significantly improving compression performance. Additionally, we introduce a Mixed Local-Global Feed Forward Network (MLGFFN) within the Transformer block to enhance the extraction of diverse and rich information, which is crucial for effective compression. These innovations collectively improve the transformation's ability to project data into a more decorrelated latent space, thereby boosting overall compression efficiency. Experimental results demonstrate that our framework surpasses state-of-the-art LIC methods in rate-distortion performance.