conference paper

Conference/ProceedingsPicture Coding Symposium (PCS 2018)
Start date24.06.2018
End date27.06.2018
AddressSan Francisco, CA, USA
Author(s)Markus Küchhold, Maik Simon, Thomas Sikora
TitleRestricted Boltzmann Machine Image Compression
AbstractWe propose a novel lossy block-based image compression approach. Our approach builds on non-linear autoencoders that can, when properly trained, explore non-linear statistical dependencies in the image blocks for redundancy reduction. In contrast the DCT employed in JPEG is inherently restricted to exploration of linear dependencies using a second-order statistics framework. The coder is based on pre-trained class-specific Restricted Boltzmann Machines (RBM). These machines are statistical variants of neural network autoencoders that directly map pixel values in image blocks into coded bits. Decoders can be implemented with low computational complexity in a codebook design. Experimental results show that our RBM-codec outperforms JPEG at high compression rates, both in terms of PSNR, SSIM and subjective results.
Key wordsRestricted Boltzmann Machines, Autoencoder, Image Compression
NoteDOI: 10.1109/PCS.2018.8456279