SFTNet: Spatial-Frequency Transformer Network for Learned Image Compression
- 3 December 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Broadcasting
- Vol. 72 (1) , 247-259
Abstract
In learned image compression (LIC), most existing methods process spatial domain information through convolutional neural networks (CNNs) or Transformers. However, they struggle to model frequency-domain correlations, particularly high-frequency detail correlations across regions. They suffer from high-frequency detail redundancy and inefficient uniform bit allocation, neither of which adapts to regional complexity. To address these issues, we propose the Spatial-Frequency Transformer Network (SFTNet) for LIC, which models cross-domain correlations between spatial structures and frequency components to enable region-adaptive bit allocation. To model these correlations, the Frequency-aware Transformer Block (FATB) employs dual attention mechanisms to collaboratively process spatial and frequency components. Specifically, its intra-block frequency attention enhances high-frequency details within each block by using low-frequency components as local structural guidance. Meanwhile, its inter-block spatial attention captures global consistency by modeling cross-block dependencies among low-frequency components. Building on this, the Feature Re-weighting Strategy (FRS) evaluates block importance via joint analysis of spatial dependencies and high-frequency energy. This enables dynamic alignment of the spatial-frequency features modeled by FATB for adaptive bit allocation, prioritizing structurally complex regions and compressing smooth areas to reduce redundancy. Experimental results demonstrate that our SFTNet outperforms VVC by −6.55% and −6.18% in BD-Rate on Kodak and CLIC datasets while achieving better reconstruction quality of the high-frequency details compared to recent LIC methods.Keywords
Funding Information
- National Natural Science Foundation of China (62372036)
- National Natural Science Foundation of China (62120106009)
- National Natural Science Foundation of China (62332017)
- National Natural Science Foundation of China (U24B20179)
- National Natural Science Foundation of China (U22A2022)
This publication has 42 references indexed in Scilit:
- Window-Based Channel Attention for Wavelet-Enhanced Learned Image Compression2024
- WeConvene: Learned Image Compression with Wavelet-Domain Convolution and Entropy Model2024
- Joint Demosaicing and Denoising with Frequency Domain Features2023
- ELIC: Efficient Learned Image Compression with Unevenly Grouped Space-Channel Contextual Adaptive Coding2022
- Learned Image Compression With Discretized Gaussian Mixture Likelihoods and Attention Modules2020
- ImageNet: A large-scale hierarchical image database2009
- A two-layered wavelet-based algorithm for efficient lossless and lossy image compressionIEEE Transactions on Circuits and Systems for Video Technology, 2000
- The JPEG still picture compression standardIEEE Transactions on Consumer Electronics, 1992
- A theory for multiresolution signal decomposition: the wavelet representationIEEE Transactions on Pattern Analysis and Machine Intelligence, 1989
- Adaptive bit allocation for image compressionComputer Graphics and Image Processing, 1982