Two-stage pyramidal convolutional neural networks for image colorization

Yu Jen Wei, Tsu Tsai Wei, Tien Ying Kuo, Po Chyi Su

Research output: Contribution to journalArticlepeer-review


The development of colorization algorithms through deep learning has become the current research trend. These algorithms colorize grayscale images automatically and quickly, but the colors produced are usually subdued and have low saturation. This research addresses this issue of existing algorithms by presenting a two-stage convolutional neural network (CNN) structure with the first and second stages being a chroma map generation network and a refinement network, respectively. To begin, we convert the color space of an image from RGB to HSV to predict its low-resolution chroma components and therefore reduce the computational complexity. Following that, the first-stage output is zoomed in and its detail is enhanced with a pyramidal CNN, resulting in a colorized image. Experiments show that, while using fewer parameters, our methodology produces results with more realistic color and higher saturation than existing methods.

Original languageEnglish
Article numbere15
JournalAPSIPA Transactions on Signal and Information Processing
StatePublished - 8 Oct 2021


  • Convolutional neural network
  • Image colorization
  • Image pyramid


Dive into the research topics of 'Two-stage pyramidal convolutional neural networks for image colorization'. Together they form a unique fingerprint.

Cite this