| Session: | Machine Learning for Image and Video Coding |
| Location: | Lecture Room |
| Session Time: | Monday, June 25, 10:20 - 12:40 |
| Presentation Time: | Monday, June 25, 10:20 - 10:40 |
| Presentation: |
Special Session Lecture
|
| Paper Title: |
RESTRICTED BOLTZMANN MACHINE IMAGE COMPRESSION |
| Authors: |
Markus Küchhold; Technische Universität Berlin, Germany | | |
| | Maik Simon; Technische Universität Berlin, Germany | | |
| | Thomas Sikora; Technische Universität Berlin, Germany | | |
| Abstract: |
We 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. |