Invertible Grayscale with Sparsity Enforcing Priors

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journalpeer-review

View graph of relations

Author(s)

  • Yong DU
  • Yangyang XU
  • Taizhong YE
  • Qiang WEN
  • Junyu DONG
  • Guoqiang HAN
  • Shengfeng HE

Related Research Unit(s)

Detail(s)

Original languageEnglish
Article number97
Number of pages17
Journal / PublicationACM Transactions on Multimedia Computing, Communications and Applications
Volume17
Issue number3
Online published22 Jul 2021
Publication statusPublished - Jul 2021

Abstract

Color dimensionality reduction is believed as a non-invertible process, as re-colorization results in perceptually noticeable and unrecoverable distortion. In this paper, we propose to convert a color image into a grayscale image that can fully recover its original colors, and more importantly, the encoded information is discriminative and sparse that save storage capacity. Particularly, we design an invertible deep neural network for color encoding and decoding purposes. This network learns to generate a residual image that encodes color information, and it is then combined with a base grayscale image for color recovering. In this way, the non-differentiable compression process (e.g., JPEG) of the base grayscale image can be integrated into the network in an end-to-end manner. To further reduce the size of the residual image, we present a specific layer to enhance Sparsity Enforcing Priors (SEP), and thus leading to the negligible storage space. The proposed method allows color embedding on a sparse residual image, while keeping a high, 35dB PSNR on average. Extensive experiments demonstrate that the proposed method outperforms state-of-the-arts in terms of image quality and tolerability to compression.

Research Area(s)

  • colorization, convolutional neural networks, Decolorization, sparsity enforcing priors

Bibliographic Note

Research Unit(s) information for this publication is provided by the author(s) concerned.

Citation Format(s)

Invertible Grayscale with Sparsity Enforcing Priors. / DU, Yong; XU, Yangyang; YE, Taizhong; WEN, Qiang; XIAO, Chufeng; DONG, Junyu; HAN, Guoqiang; HE, Shengfeng.

In: ACM Transactions on Multimedia Computing, Communications and Applications, Vol. 17, No. 3, 97, 07.2021.

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journalpeer-review