A Novel Image Cartoonization Algorithm without Deep Learning

Konferenz: CIBDA 2022 - 3rd International Conference on Computer Information and Big Data Applications
25.03.2022 - 27.03.2022 in Wuhan, China

Tagungsband: CIBDA 2022

Seiten: 4Sprache: EnglischTyp: PDF

Autoren:
Wang, Shanshan (Wuhan Polytechnic, Wuhan, China)
Qi, Jiangyuan (Changjiang Polytechnic, China)

Inhalt:
In this paper, we propose an algorithm to transforming photos of real-world scenes into cartoon style images, which is valuable and challenging in computer vision and computer graphics. Our solution belongs to traditional image processing-based methods, which have recently ignored to stylize images in artistic forms such as painting while deep learning has become popular. However, learning methods produce cartoonization by consuming time, quantity data and energy. Due to the fact that cartoon styles have similar characteristics with simplification and abstraction, and cartoon images tend to have clear edges, smooth color shading and relatively simple textures, which is quite make a mountain out of a molehill for learning-based algorithm used in existing methods. In this paper, we propose a novel image cartoonization algorithm without deep learning. Our method does not need prepare dataset and training, which is easy to use. Our method is also producing satisfied vision results. Experimental results show that our method is able to generate high-quality cartoon images from real-world photos and outperforms state-of-the-art methods.