Learned image compression (LIC) has gained traction as an effective solution
for image storage and transmission in recent years. However, existing LIC
methods are redundant in latent representation due to limitations in capturing
anisotropic frequency components and preserving directional details. To
overcome these challenges, we propose a novel frequency-aware transformer (FAT)
block that for the first time achieves multiscale directional ananlysis for
LIC. The FAT block comprises frequency-decomposition window attention (FDWA)
modules to capture multiscale and directional frequency components of natural
images. Additionally, we introduce frequency-modulation feed-forward network
(FMFFN) to adaptively modulate different frequency components, improving
rate-distortion performance. Furthermore, we present a transformer-based
channel-wise autoregressive (T-CA) model that effectively exploits channel
dependencies. Experiments show that our method achieves state-of-the-art
rate-distortion performance compared to existing LIC methods, and evidently
outperforms latest standardized codec VTM-12.1 by 14.5%, 15.1%, 13.0% in
BD-rate on the Kodak, Tecnick, and CLIC datasets.

学习图像压缩 (LIC) 已成为近年来在图像存储和传输方面的有效解决方案，然而，现有的 LIC 方法存在潜在的表示冗余问题，限制了其对各向异性频率成分和方向细节的捕捉。为了克服这些挑战，我们提出了一种新颖的频率感知变换器 (FAT) 模块，首次实现了对 LIC 的多尺度方向分析。FAT 模块由频率分解窗口注意力 (FDWA) 模块组成，用于捕捉自然图像的多尺度和方向频率成分。此外，我们引入了频率调制前馈网络 (FMFFN)，以自适应调制不同频率成分，从而改善了速率失真性能。此外，我们还提出了基于变换器的通道级自回归 (T-CA) 模型，有效地利用了通道依赖性。实验证明，与现有的 LIC 方法相比，我们的方法在速率失真性能上达到了最先进的水平，并在 Kodak、Tecnick 和 CLIC 数据集上相对于最新的标准编解码器 VTM-12.1 提高了 14.5%、15.1% 和 13.0% 的 BD - 率。