Window-based Channel Attention for Wavelet-enhanced Learned Image Compression

TL;DR

This paper introduces a novel window-based channel attention mechanism combined with wavelet transforms in a hybrid framework, significantly enhancing receptive fields and global information modeling for learned image compression, leading to state-of-the-art results.

Contribution

It proposes a new window-based channel attention method and integrates wavelet transforms into LIC, enabling larger receptive fields and improved global correlation modeling.

Findings

Achieves state-of-the-art rate-distortion performance.

Reduces BD-rate by up to 24.71% on standard datasets.

Enlarges receptive fields for better large object modeling.

Abstract

Learned Image Compression (LIC) models have achieved superior rate-distortion performance than traditional codecs. Existing LIC models use CNN, Transformer, or Mixed CNN-Transformer as basic blocks. However, limited by the shifted window attention, Swin-Transformer-based LIC exhibits a restricted growth of receptive fields, affecting the ability to model large objects for image compression. To address this issue and improve the performance, we incorporate window partition into channel attention for the first time to obtain large receptive fields and capture more global information. Since channel attention hinders local information learning, it is important to extend existing attention mechanisms in Transformer codecs to the space-channel attention to establish multiple receptive fields, being able to capture global correlations with large receptive fields while maintaining detailed characterization of local correlations with small receptive fields. We also incorporate the discrete wavelet transform into our Spatial-Channel Hybrid (SCH) framework for efficient frequency-dependent down-sampling and further enlarging receptive fields. Experiment results demonstrate that our method achieves state-of-the-art performances, reducing BD-rate by 18.54%, 23.98%, 22.33%, and 24.71% on four standard datasets compared to VTM-23.1.