We propose Quality Enhancement via a Side bitstream Network (QESN) technique for lossy image compression. The proposed QESN utilizes the network architecture of deep image compression to produce a bitstream for enhancing the quality of conventional compression. We also present a loss function that directly optimizes the Bjontegaard delta bit rate (BD-BR) by using a differentiable model of a rate-distortion curve. Experimental results show that QESN improves the rate by 16.7% in the BD-BR compared to Better Portable Graphics.
Takahiro NARUKO
Hitachi, Ltd.
Hiroaki AKUTSU
Hitachi, Ltd.
Koki TSUBOTA
The University of Tokyo
Kiyoharu AIZAWA
The University of Tokyo
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Takahiro NARUKO, Hiroaki AKUTSU, Koki TSUBOTA, Kiyoharu AIZAWA, "Quality Enhancement of Conventional Compression with a Learned Side Bitstream" in IEICE TRANSACTIONS on Information,
vol. E106-D, no. 8, pp. 1296-1299, August 2023, doi: 10.1587/transinf.2022EDL8096.
Abstract: We propose Quality Enhancement via a Side bitstream Network (QESN) technique for lossy image compression. The proposed QESN utilizes the network architecture of deep image compression to produce a bitstream for enhancing the quality of conventional compression. We also present a loss function that directly optimizes the Bjontegaard delta bit rate (BD-BR) by using a differentiable model of a rate-distortion curve. Experimental results show that QESN improves the rate by 16.7% in the BD-BR compared to Better Portable Graphics.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2022EDL8096/_p
Copy
@ARTICLE{e106-d_8_1296,
author={Takahiro NARUKO, Hiroaki AKUTSU, Koki TSUBOTA, Kiyoharu AIZAWA, },
journal={IEICE TRANSACTIONS on Information},
title={Quality Enhancement of Conventional Compression with a Learned Side Bitstream},
year={2023},
volume={E106-D},
number={8},
pages={1296-1299},
abstract={We propose Quality Enhancement via a Side bitstream Network (QESN) technique for lossy image compression. The proposed QESN utilizes the network architecture of deep image compression to produce a bitstream for enhancing the quality of conventional compression. We also present a loss function that directly optimizes the Bjontegaard delta bit rate (BD-BR) by using a differentiable model of a rate-distortion curve. Experimental results show that QESN improves the rate by 16.7% in the BD-BR compared to Better Portable Graphics.},
keywords={},
doi={10.1587/transinf.2022EDL8096},
ISSN={1745-1361},
month={August},}
Copy
TY - JOUR
TI - Quality Enhancement of Conventional Compression with a Learned Side Bitstream
T2 - IEICE TRANSACTIONS on Information
SP - 1296
EP - 1299
AU - Takahiro NARUKO
AU - Hiroaki AKUTSU
AU - Koki TSUBOTA
AU - Kiyoharu AIZAWA
PY - 2023
DO - 10.1587/transinf.2022EDL8096
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E106-D
IS - 8
JA - IEICE TRANSACTIONS on Information
Y1 - August 2023
AB - We propose Quality Enhancement via a Side bitstream Network (QESN) technique for lossy image compression. The proposed QESN utilizes the network architecture of deep image compression to produce a bitstream for enhancing the quality of conventional compression. We also present a loss function that directly optimizes the Bjontegaard delta bit rate (BD-BR) by using a differentiable model of a rate-distortion curve. Experimental results show that QESN improves the rate by 16.7% in the BD-BR compared to Better Portable Graphics.
ER -