Abstract
It has become a trend in recent years to use deep neural networks for colorization. However, previous methods often encounter problems with edge color leakage and difficulties in obtaining a plausible color output from the Euclidean distance. To solve these problems, we propose a new adversarial edge-aware image colorization method with multitask output combined with semantic segmentation. The system uses a generator with a deep semantic fusion structure to infer semantic clues in a given grayscale image under chroma conditions and learns colorization by simultaneously predicting color information and semantic information. In addition, we also use a specific color difference loss with characteristics of human visual observation that is combined with semantic segmentation loss and adversarial loss for training. The experimental results show that our method is superior to existing methods in terms of different quality metrics and achieves good results in image colorization.
Highlights
In regard to coloring black and white photos, the first thought that comes to mind is the work of an artist named Marina Amaral
The most mentioned methods are [11]–[17], [35]. These datadriven methods all have good performance, we find that the existing models do not substantively take into account the effective colorization of the object edge in the image and are limited in the selection of the loss function
We propose a new framework of an edgeaware colorized deep neural network with semantic segmentation to solve the above problems
Summary
In regard to coloring black and white photos, the first thought that comes to mind is the work of an artist named Marina Amaral. She used postprocessing to fill in the color of many famous historical photos, and the works were realistic and did not contain any holes. Users marked colors on the gray image in different areas and colored the image through local diffusion. These methods [1]–[4] require the user to draw colored strokes on a grayscale image. An optimization program generates a color image to match the user’s scribble
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have