Corpus ID: 850237

Soft-to-Hard Vector Quantization for End-to-End Learning Compressible Representations

@inproceedings{Agustsson2017SofttoHardVQ,
  title={Soft-to-Hard Vector Quantization for End-to-End Learning Compressible Representations},
  author={Eirikur Agustsson and Fabian Mentzer and Michael Tschannen and Lukas Cavigelli and Radu Timofte and Luca Benini and Luc Van Gool},
  booktitle={NIPS},
  year={2017}
}
We present a new approach to learn compressible representations in deep architectures with an end-to-end training strategy. Our method is based on a soft (continuous) relaxation of quantization and entropy, which we anneal to their discrete counterparts throughout training. We showcase this method for two challenging applications: Image compression and neural network compression. While these tasks have typically been approached with different methods, our soft-to-hard quantization approach… Expand
Deep Learning-Based Image Compression with Trellis Coded Quantization
TLDR
This paper develops a simple image compression model that consists of three subnetworks (encoder, decoder and entropy estimation), and optimize all of the components in an end-to-end manner and shows the advantage of TCQ. Expand
Ensemble Learning-Based Rate-Distortion Optimization for End-to-End Image Compression
TLDR
This work proposes several methods to obtain multiple models for end-to-end image compression using trained deep networks as encoding/decoding models as an ensemble, and proposes a geometric transform-based self-ensemble method. Expand
Deep Generative Video Compression
TLDR
This work proposes an end-to-end, deep generative modeling approach to compress temporal sequences with a focus on video that builds upon variational autoencoder models for sequential data and combines them with recent work on neural image compression. Expand
Deep Generative Video Compression
The usage of deep generative models for image compression has led to impressive performance gains over classical codecs while neural video compression is still in its infancy. Here, we propose anExpand
Deep Generative Video Compression
The usage of deep generative models for image compression has led to impressive performance gains over classical codecs while neural video compression is still in its infancy. Here, we propose anExpand
An End-to-End Learning Framework 2 for Video Compression
5 Abstract—Traditional video compression approaches build upon the hybrid coding framework with motion-compensated prediction 6 and residual transform coding. In this paper, we propose the firstExpand
Learned transform compression with optimized entropy encoding
TLDR
A soft relaxation of the quantization operation is utilized to allow for back-propagation of gradients and employ vector (rather than scalar) quantization of the latent codes enabling direct optimization of the code entropy. Expand
Conditional Probability Models for Deep Image Compression
TLDR
This paper proposes a new technique to navigate the rate-distortion trade-off for an image compression auto-encoder by using a context model: A 3D-CNN which learns a conditional probability model of the latent distribution of the auto- Encoder. Expand
Universal Deep Neural Network Compression
TLDR
This work for the first time introduces universal DNN compression by universal vector quantization and universal source coding, which utilizes universal lattice quantization, which randomizes the source by uniform random dithering before latticequantization and can perform near-optimally on any source without relying on knowledge of the source distribution. Expand
Rate Distortion Characteristic Modeling for Neural Image Compression
TLDR
This paper proposes a plugin-in module to learn the relationship between the target bit-rate and the binary representation for the latent variable of auto-encoder, and model the rate and distortion characteristic of NIC as a function of the coding parameter λ respectively. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 46 REFERENCES
Soft Weight-Sharing for Neural Network Compression
TLDR
This paper shows that competitive compression rates can be achieved by using a version of  “soft weight-sharing” (Nowlan & Hinton, 1992) and achieves both quantization and pruning in one simple (re-)training procedure, exposing the relation between compression and the minimum description length (MDL) principle. Expand
Incremental Network Quantization: Towards Lossless CNNs with Low-Precision Weights
TLDR
Extensive experiments on the ImageNet classification task using almost all known deep CNN architectures including AlexNet, VGG-16, GoogleNet and ResNets well testify the efficacy of the proposed INQ, showing that at 5-bit quantization, models have improved accuracy than the 32-bit floating-point references. Expand
End-to-end optimization of nonlinear transform codes for perceptual quality
TLDR
This work introduces a general framework for end-to-end optimization of the rate-distortion performance of nonlinear transform codes assuming scalar quantization and considers a code built from a linear transform followed by a form of multi-dimensional local gain control. Expand
Deep Compression: Compressing Deep Neural Network with Pruning, Trained Quantization and Huffman Coding
TLDR
This work introduces "deep compression", a three stage pipeline: pruning, trained quantization and Huffman coding, that work together to reduce the storage requirement of neural networks by 35x to 49x without affecting their accuracy. Expand
Lossy Image Compression with Compressive Autoencoders
TLDR
It is shown that minimal changes to the loss are sufficient to train deep autoencoders competitive with JPEG 2000 and outperforming recently proposed approaches based on RNNs, and furthermore computationally efficient thanks to a sub-pixel architecture, which makes it suitable for high-resolution images. Expand
End-to-end Optimized Image Compression
TLDR
Across an independent set of test images, it is found that the optimized method generally exhibits better rate-distortion performance than the standard JPEG and JPEG 2000 compression methods, and a dramatic improvement in visual quality is observed, supported by objective quality estimates using MS-SSIM. Expand
Towards the Limit of Network Quantization
TLDR
It is derived that the network quantization problem can be related to the entropy-constrained scalar quantization (ECSQ) problem in information theory and two solutions of ECSQ are proposed, i.e., uniform quantization and an iterative solution similar to Lloyd's algorithm. Expand
Full Resolution Image Compression with Recurrent Neural Networks
TLDR
This is the first neural network architecture that is able to outperform JPEG at image compression across most bitrates on the rate-distortion curve on the Kodak dataset images, with and without the aid of entropy coding. Expand
Using very deep autoencoders for content-based image retrieval
TLDR
This work shows how to learn many layers of features on color images and how these features are used to initialize deep autoencoders, which are then used to map images to short binary codes. Expand
Learning Structured Sparsity in Deep Neural Networks
TLDR
The results show that for CIFAR-10, regularization on layer depth can reduce 20 layers of a Deep Residual Network to 18 layers while improve the accuracy from 91.25% to 92.60%, which is still slightly higher than that of original ResNet with 32 layers. Expand
...
1
2
3
4
5
...