Fetching the paper…

Reading the bibliography…

2017

Image Compression: Sparse Coding vs. Bottleneck Autoencoders

Watkins, Yijing, Sayeh, Mohammad, Iaroshenko, Oleksandr et al.

Original
PDF
Post on X

Understand

Bottleneck autoencoders have been actively researched as a solution to image compression tasks.

However, we observed that bottleneck autoencoders produce subjectively low quality reconstructed images.
In this work, we explore the ability of sparse coding to improve reconstructed image quality for the same degree of compression.
We observe that sparse image compression produces visually superior reconstructed images and yields higher values of pixel-wise measures of reconstruction quality (PSNR and SSIM) compared to bottleneck autoencoders.

Image Compression: Sparse Coding vs. Bottleneck Autoencoders · Around

Built on

Compressed sensing
D. L. Donoho · 2006
Earlier work this paper cites.
Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information
E. J. C. et al · 2006
Earlier work this paper cites.
Reducing the dimensionality of data with neural networks
G. H. et al · 2006
Earlier work this paper cites.
Sparse coding via thresholding and local competition in neural circuits
C. J. R. et al · 2008
Earlier work this paper cites.
Deep boltzmann machines
R. S. et al · 2009
Earlier work this paper cites.

Similar

Learning multiple layers of features from tiny images
A. Krizhevsky · 2009
Cited alongside, same era.
Adaptive compressive sensing for low power wireless sensors
A. W. et al · 2014
Cited alongside, same era.
Software available from tensorflow.org
TensorFlow: Large-scale machine learning on heterogeneous systems, 2015 · 2015
Cited alongside, same era.
Variable rate image compression with recurrent neural networks
Original
G. T. et al · 2015
Cited alongside, same era.
Software available from petavision.github.io
Petavision
Cited in the paper.

Then

Generating images with perceptual similarity metrics based on deep networks
A. D. et al · 2016
Later among the works it cites.
Perceptual losses for real-time style transfer and super-resolution
J. J. et al · 2016
Later among the works it cites.
Image data compression and noisy channel error correction using deep neural network
Y. W. et al · 2016
Later among the works it cites.
EnhanceNet: Single image super-resolution through automated texture synthesis
S. et al · 2017
Closest in time.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…