Non-local Attention Optimized Deep Image Compression
This work addresses image compression for applications requiring high-quality storage or transmission, representing an incremental improvement over existing deep learning and conventional methods.
The paper tackles image compression by proposing a Non-Local Attention Optimized Deep Image Compression (NLAIC) framework that embeds non-local operations and attention mechanisms to capture correlations and adapt bit allocation, resulting in outperforming existing methods on the Kodak dataset for PSNR and MS-SSIM metrics.
This paper proposes a novel Non-Local Attention Optimized Deep Image Compression (NLAIC) framework, which is built on top of the popular variational auto-encoder (VAE) structure. Our NLAIC framework embeds non-local operations in the encoders and decoders for both image and latent feature probability information (known as hyperprior) to capture both local and global correlations, and apply attention mechanism to generate masks that are used to weigh the features for the image and hyperprior, which implicitly adapt bit allocation for different features based on their importance. Furthermore, both hyperpriors and spatial-channel neighbors of the latent features are used to improve entropy coding. The proposed model outperforms the existing methods on Kodak dataset, including learned (e.g., Balle2019, Balle2018) and conventional (e.g., BPG, JPEG2000, JPEG) image compression methods, for both PSNR and MS-SSIM distortion metrics.