Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Sep 30;26(10):836.
doi: 10.3390/e26100836.

Information Bottleneck Driven Deep Video Compression-IBOpenDVCW

Timor Leiderman  1 , Yosef Ben Ezra  1
Affiliations

Information Bottleneck Driven Deep Video Compression-IBOpenDVCW

Timor Leiderman et al. Entropy (Basel). .

Abstract

Video compression remains a challenging task despite significant advancements in end-to-end optimized deep networks for video coding. This study, inspired by information bottleneck (IB) theory, introduces a novel approach that combines IB theory with wavelet transform. We perform a comprehensive analysis of information and mutual information across various mother wavelets and decomposition levels. Additionally, we replace the conventional average pooling layers with a discrete wavelet transform creating more advanced pooling methods to investigate their effects on information and mutual information. Our results demonstrate that the proposed model and training technique outperform existing state-of-the-art video compression methods, delivering competitive rate-distortion performance compared to the AVC/H.264 and HEVC/H.265 codecs.

Keywords: deep video compression; information bottleneck; neural networks; wavelets.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflicts of interest.

Figures

Figure 1
Figure 1
High-Level framework of the OpenDVCW Network.
Figure 2
Figure 2
Pyramid architecture of the optical flow estimation.
Figure 3
Figure 3
Visualization of DWT as we apply the transform on the approximation on every iteration.
Figure 4
Figure 4
Calculated information on Lenna image for various mother wavelets.
Figure 5
Figure 5
Performance on the UVG dataset comparison between AVC/H.264, HEVC/H.265, VVC/H.266 and OpenDVCW with Db2, Sym3 and Haar wavelets for the DWT transform in the optical flow. (A)—Beauty, (B)—HoneyBee, (C)—ShakeNDry, (D)—Bosphorus.
Figure 5
Figure 5
Performance on the UVG dataset comparison between AVC/H.264, HEVC/H.265, VVC/H.266 and OpenDVCW with Db2, Sym3 and Haar wavelets for the DWT transform in the optical flow. (A)—Beauty, (B)—HoneyBee, (C)—ShakeNDry, (D)—Bosphorus.
See this image and copyright information in PMC

References

    1. Symes P. Digital Video Compression. McGraw-Hill; New York, NY, USA: 2004. (Digital Video/Audio Series).
    1. Zhang Z., Shi Y., Toda H., Akiduki T. A Study of a new wavelet neural network for deep learning; Proceedings of the 2017 International Conference on Wavelet Analysis and Pattern Recognition (ICWAPR); Ningbo, China. 9–12 July 2017; pp. 127–131. - DOI
    1. Ma S., Zhang X., Jia C., Zhao Z., Wang S., Wang S. Image and Video Compression with Neural Networks: A Review. arXiv. 2019 doi: 10.1109/TCSVT.2019.2910119.1904.03567 - DOI
    1. Shin S. Industrial application of wavelet analysis; Proceedings of the 2008 International Conference on Wavelet Analysis and Pattern Recognition; Hong Kong, China. 30–31 August 2008; pp. 607–610. - DOI
    1. Keinert F. Wavelets and Multiwavelets. Studies in Advanced Mathematics; CRC Press; Boca Raton, FL, USA: 2003.

LinkOut - more resources