Research map / Image generation / Editing and style transfer
Learning to Cartoonize Using White-Box Cartoon Representations
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| decompose images into surface, structure, and texture representations and train with separate objectives | image cartoonization · Scenery | FID to Cartoon: 130.38 -> 101.31 | Helped |
| decompose images into surface, structure, and texture representations and train with separate objectives | image cartoonization · Scenery | FID to Photo: 75.28 -> 28.79 | Helped |
| use the white-box representation-guided GAN framework | image cartoonization · User study | cartoon quality mean score: 2.977 -> 4.017 | Helped |
| use the white-box representation-guided GAN framework | image cartoonization · User study | overall quality mean score: 2.38 -> 3.877 | Helped |
| use the white-box cartoonization model | image cartoonization · LR 256*256, CPU | inference time (ms): 639.31 -> 64.66 | Helped |
| use the white-box cartoonization model | image cartoonization · LR 256*256, GPU | inference time (ms): 9.22 -> 3.58 | Helped |
| use the white-box cartoonization model | image cartoonization · HR 720*1280, GPU | inference time (ms): 48.96 -> 17.23 | Helped |
| use the white-box cartoonization model | image cartoonization · model | parameters (M): 1.68 -> 1.48 | Helped |
| add the texture representation (ablation) | full cartoonization model · cartoonization | detail cleanliness | Helped |
| add the structure representation (ablation) | full cartoonization model · cartoonization | high-frequency noise | Helped |
| add the surface representation (ablation) | full cartoonization model · cartoonization | surface smoothness | Helped |
| increase the texture-representation loss weight | cartoonization model · style control | retained high-frequency detail | Helped |
Closest papers by meaning
- CartoonGAN: Generative Adversarial Networks for Photo Cartoonization (CVPR 2018 · similarity 0.92)
- Learning to Incorporate Texture Saliency Adaptive Attention to Image Cartoonization (ICML 2022 · similarity 0.89)
- Interactive Cartoonization With Controllable Perceptual Factors (CVPR 2023 · similarity 0.86)
- A Learned Representation For Artistic Style (ICLR 2017 · similarity 0.85)
- Stylized Neural Painting (CVPR 2021 · similarity 0.85)
- APDrawingGAN: Generating Artistic Portrait Drawings From Face Photos With Hierarchical GANs (CVPR 2019 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.