Attention-guided Image Compression by Deep Reconstruction of Compressive Sensed Saliency Skeleton

Abstract

We propose a deep learning system for attention-guided dual-layer image compression (AGDL). In the AGDL compression system, an image is encoded into two layers, a base layer and an attention-guided refinement layer. Unlike the existing ROI image compression methods that spend an extra bit budget equally on all pixels in ROI, AGDL employs a CNN module to predict those pixels on and near a saliency sketch within ROI that are critical to perceptual quality. Only the critical pixels are further sampled by compressive sensing (CS) to form a very compact refinement layer. Another novel CNN method is developed to jointly decode the two compression layers for a much refined reconstruction, while strictly satisfying the transmitted CS constraints on perceptually critical pixels. Extensive experiments demonstrate that the proposed AGDL system advances the state of the art in perception-aware image compression.

arXiv commentAccepted to CVPR 2021

References in corpus (12)

• Adam: A Method for Stochastic Optimization
• Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
• Variational image compression with a scale hyperprior
• End-to-end Optimized Image Compression
• Joint Autoregressive and Hierarchical Priors for Learned Image Compression
• Lossy Image Compression with Compressive Autoencoders
• Variable Rate Image Compression with Recurrent Neural Networks
• Real-Time Adaptive Image Compression
• Implicit Dual-domain Convolutional Network for Robust Color Image Compression Artifact Reduction
• Towards High-Resolution Salient Object Detection
• One-to-Many Network for Visually Pleasing Compression Artifacts Reduction
• Towards Real-Time Automatic Portrait Matting on Mobile Devices