Block-optimized Variable Bit Rate Neural Image Compression

05/28/2018

∙

In this work, we propose an end-to-end block-based auto-encoder system for image compression. We introduce novel contributions to neural-network based image compression, mainly in achieving binarization simulation, variable bit rates with multiple networks, entropy-friendly representations, inference-stage code optimization and performance-improving normalization layers in the auto-encoder. We evaluate and show the incremental performance increase of each of our contributions.

READ FULL TEXT

Block-optimized Variable Bit Rate Neural Image Compression

Layered Image Compression using Scalable Auto-encoder

End-to-End Learned Block-Based Image Compression with Block-Level Masked Convolutions and Asymptotic Closed Loop Training

Conditional Probability Models for Deep Image Compression

Variable Rate Video Compression using a Hybrid Recurrent Convolutional Learning Framework

MBRS : Enhancing Robustness of DNN-based Watermarking by Mini-Batch of Real and Simulated JPEG Compression

Generalization Gap in Amortized Inference

Orthogonal Features-based EEG Signal Denoising using Fractionally Compressed AutoEncoder

Block-optimized Variable Bit Rate Neural Image Compression

Related Research

Layered Image Compression using Scalable Auto-encoder

End-to-End Learned Block-Based Image Compression with Block-Level Masked Convolutions and Asymptotic Closed Loop Training

Conditional Probability Models for Deep Image Compression

Variable Rate Video Compression using a Hybrid Recurrent Convolutional Learning Framework

MBRS : Enhancing Robustness of DNN-based Watermarking by Mini-Batch of Real and Simulated JPEG Compression

Generalization Gap in Amortized Inference

Orthogonal Features-based EEG Signal Denoising using Fractionally Compressed AutoEncoder