Cartoon style generation of animated image scenes based on improved GAN and RetNet
Zhenyao Jin, Lei Yang · 2026
Read Cartoon style generation of animated image scenes based on improved GAN and RetNet by Zhenyao Jin, Lei Yang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0347014.txt.
Description
Zhenyao Jin and Lei Yang present a system that turns ordinary images into consistent cartoon scenes, in PLOS ONE in 2026. Automated cartoon making often drifts in style from frame to frame or warps shapes, which limits its use in film and games. Their model joins an improved adversarial network, where two programs compete to sharpen output, with a retention network that holds onto structure across the image.
Extra parts handle style variety, judging, and geometric consistency, including a module that keeps surfaces from bending wrongly. Tested on two public scene datasets against three mainstream rivals, the model scored well on image quality, structure preservation, speed, and modest memory use. Its cartoon output stayed close to the original layout. The authors pitch it as a cheaper automated route for animation, game art, and digital content production.
Catalog record
| CREATOR | Zhenyao Jin, Lei Yang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 61779 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |