Research map / Image generation / Editing and style transfer
Deep Plastic Surgery: Robust and Controllable Image Editing with Human-Drawn Sketches
Shuai Yang, Zhangyang Wang, Jiaying Liu, Zongming Guo
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| feed refined sketches (without fine-tuning) to SC-FEGAN (ablation) | SC-FEGAN · face editing | visual quality (qualitative) | Improved |
| add line deformation and discarding to dilation rough-sketch model (ablation) | Deep Plastic Surgery · CelebA-HQ sketch-to-image | structure quality (qualitative) | Improved |
| learn multiple refinement levels in one model instead of one level per model (ablation) | Deep Plastic Surgery · CelebA-HQ sketch-to-image | visual quality (qualitative) | Improved |
| add adaptation to fixed edge model F (I_out loss terms) (ablation) | Deep Plastic Surgery · CelebA-HQ sketch-to-image | line distinctness (qualitative) | Improved |
| style-based (AdaIN) conditioning instead of label concatenation or controllable resblock (ablation) | Deep Plastic Surgery · CelebA-HQ refinement level control | visual quality (qualitative) | Improved |
| use feed-forward model instead of ContextualGAN nearest-neighbor search | face synthesis · 64x64 CelebA | time per image: 7.89 s -> 12 ms | Improved |
| use Deep Plastic Surgery instead of SC-FEGAN | face editing · Face editing user study | mean preference ratio: 0.238 -> 0.730 | Improved |
| use Deep Plastic Surgery instead of pix2pixHD | face synthesis · Face synthesis user study | mean preference ratio: 0.031 -> 0.945 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- SC-FEGAN — SC-FEGAN: Face Editing Generative Adversarial Network With User's Sketch and Color (ICCV 2019) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- SketchEdit: Mask-Free Local Image Manipulation With Partial Sketches (CVPR 2022 · similarity 0.89)
- It's All About Your Sketch: Democratising Sketch Control in Diffusion Models (CVPR 2024 · similarity 0.86)
- Sketch Down the FLOPs: Towards Efficient Networks for Human Sketch (CVPR 2025 · similarity 0.85)
- PIXELS: Progressive Image Xemplar-based Editing with Latent Surgery (AAAI 2025 · similarity 0.85)
- One-Shot Reference-based Structure-Aware Image to Sketch Synthesis (AAAI 2025 · similarity 0.85)
- Image Sculpting: Precise Object Editing with 3D Geometry Control (CVPR 2024 · 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.
