A Multi-Style Image and Video Cartoonization Framework using Deep Learning and Adaptive Style Filters

1st Muriki Parvathi
Computer Science and Engineering Rajiv Gandhi University of Knowledge Technologies
Basar, India murikiparvathi05@gmail.com
3rd Paka Akshaya
Computer Science and EngineeringRajive Gandhi University of Knowledge Technologies
Basar, India akshayapaka2003@gmail.com

2nd Dheekonda Rishitha
Computer Science and EngineeringRajive Gandhi University of Knowledge Technologies
Basar, India rishithadheekonda3@gmail.com
4th Assistant Professor Ms.Lingavva
Computer Science and EngineeringRajive Gandhi University of Knowledge Technologies
Basar, India bhanu.lingava@gmail.com

Abstract—Cartoonization of images and videos has gained sig-nificant attention due to its applications in entertainment, social media, and digital content creation. Most existing approaches fo-cus on generating a single artistic style, limiting their adaptability. In this paper, we propose a multi-style cartoonization framework that combines deep learning-based cartoonization with adaptive style filters to produce diverse cartoon effects. The system uses a white-box deep learning model for base cartoon generation and enhances the output with multiple styles such as anime, comic, and sketch. The framework supports both image and video inputs and is implemented as a real-time web-based application. Experimental results show that the proposed system generates visually appealing outputs while maintaining efficiency. Index Terms—component, formatting, style, styling, insert