What is it about?
We focus on using pix2pix, or image-to-image translation, to transfer font styles from the existing English characters set to the Khmer characters set. We collected and investigated 15 calligraphy fonts that comprise Khmer and English characters. We used the Structure Similarity Index Measure (SSIM) to assess the performance of generated images compared to ground truth images.
Featured Image
Photo by Brands&People on Unsplash
Why is it important?
The objective of this study is to support Khmer font designers in creating new Khmer font styles that align with existing English font styles. Additionally, our experiment includes the transfer of calligraphy styles from English to Khmer. By leveraging deep learning techniques, the font design process can be accelerated as algorithms can generate Khmer calligraphy that matches the font style of the English calligraphy.
Perspectives
The results show that pix2pix is capable of generating Khmer calligraphy from a simple line by learning the style from English calligraphy, aligning with our expectations.
Chanarin HENG
Kasetsart University
Read the Original
This page is a summary of: Khmer Calligraphy Style Transfer Using Pix2Pix, November 2023, Institute of Electrical & Electronics Engineers (IEEE),
DOI: 10.1109/iotais60147.2023.10346040.
You can read the full text:
Contributors
The following have contributed to this page
