Real-Time Adaptive Image Compression

Abstract

We present a machine learning-based approach to lossy image compression which outperforms all existing codecs, while running in real-time. Our algorithm typically produces files 2.5 times smaller than JPEG and JPEG 2000, 2 times smaller than WebP, and 1.7 times smaller than BPG on datasets of generic images across all quality levels. At the same time, our codec is designed to be lightweight and deployable: for example, it can encode or decode the Kodak dataset in around 10ms per image on GPU. Our architecture is an autoencoder featuring pyramidal analysis, an adaptive coding module, and regularization of the expected codelength. We also supplement our approach with adversarial training specialized towards use in a compression setting: this enables us to produce visually pleasing reconstructions for very low bitrates.

Published at ICML 2017

References in corpus (9)

• Adam: A Method for Stochastic Optimization
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks
• YFCC100M: The New Data in Multimedia Research
• Improved Techniques for Training GANs
• End-to-end Optimized Image Compression
• Lossy Image Compression with Compressive Autoencoders
• Variable Rate Image Compression with Recurrent Neural Networks
• Improved Lossy Image Compression with Priming and Spatially Adaptive Bit Rates for Recurrent Networks

Cited by in corpus (42)

• Variational image compression with a scale hyperprior
• Image and Video Compression with Neural Networks: A Review
• Joint Autoregressive and Hierarchical Priors for Learned Image Compression
• Learning for Video Compression
• Low Bit-Rate Speech Coding with VQ-VAE and a WaveNet Decoder
• Towards Image Understanding from Deep Compression without Decoding
• Fixing a Broken ELBO
• Learning to Inpaint for Image Compression
• Video Compression through Image Interpolation
• Deep Perceptual Compression
• Human Perceptual Evaluations for Image Compression
• A Provably Convergent Scheme for Compressive Sensing under Random Generative Priors
• Deep Generative Models for Distribution-Preserving Lossy Compression
• BINet: a binary inpainting network for deep patch-based image compression
• Performance Comparison of Convolutional AutoEncoders, Generative Adversarial Networks and Super-Resolution for Image Compression
• A Cross Channel Context Model for Latents in Deep Image Compression
• Neural Video Compression using Spatio-Temporal Priors
• GAN- vs. JPEG2000 Image Compression for Distributed Automotive Perception: Higher Peak SNR Does Not Mean Better Semantic Segmentation
• Recognition-Aware Learned Image Compression
• Gated Context Model with Embedded Priors for Deep Image Compression
• Validation and parameterization of a novel physics-constrained neural dynamics model applied to turbulent fluid flow
• DSSLIC: Deep Semantic Segmentation-based Layered Image Compression
• SDIT: Scalable and Diverse Cross-domain Image Translation
• Fully Convolutional Model for Variable Bit Length and Lossy High Density Compression of Mammograms
• No Multiplication? No Floating Point? No Problem! Training Networks for Efficient Inference
• CAE-ADMM: Implicit Bitrate Optimization via ADMM-based Pruning in Compressive Autoencoders
• CompressNet: Generative Compression at Extremely Low Bitrates
• Channel-Level Variable Quantization Network for Deep Image Compression
• Extreme Image Coding via Multiscale Autoencoders With Generative Adversarial Optimization
• End-to-end Optimized Video Compression with MV-Residual Prediction
• Neural Multi-scale Image Compression
• Learned Scalable Image Compression with Bidirectional Context Disentanglement Network
• Learning Compressible 360° Video Isomers
• Substitutional Neural Image Compression
• Saliency Driven Perceptual Image Compression
• Leveraging Domain Knowledge using Machine Learning for Image Compression in Internet-of-Things
• DeepSIC: Deep Semantic Image Compression
• Attention-guided Image Compression by Deep Reconstruction of Compressive Sensed Saliency Skeleton
• Principal Bit Analysis: Autoencoding with Schur-Concave Loss
• Learning to Localize Through Compressed Binary Maps
• Deep motion estimation for parallel inter-frame prediction in video compression
• Object Detection-Based Variable Quantization Processing